The mouse floated in the air in the middle of its cage.

It was an ordinary lab mouse, with white fur, beady red eyes, and an adorable little pink nose. The left ear was tattooed with a constellation of dots that identified it uniquely from all its siblings and cousins. A small nick in its tail showed where I’d taken a tissue sample soon after it had been born. The fact that it was alive today, a fully grown thirty-one-gram adult, showed that it had passed that test: it carried a modification in CHRNB2, the cholinergic receptor nicotinic beta 2 subunit gene, which we hoped would positively affect its ability to socialize with and learn from its peers.

The mouse looked at me. I looked back.

It clutched a shiny blue marble in its front paws.

It wasn’t a surprise that the mouse was floating peacefully in midair. Mice adapt quite well to zero gravity, and this particular mouse had been bred, born, and raised here on Lasky Station, as had generations of its forebears. Foremice. Predecessors.

As for myself, I floated on my side of the glass with my hands folded behind my back, my face carefully neutral. Not that it mattered directly, given that I was wearing a surgical mask, goggles, and face shield, but we’d all learned as undergrads that controlling your expression is a quick shortcut to controlling your body language and pheromones, and you do not want to influence your lab subjects in any way.

But I really, really, wanted it to put the marble in the black ring.

This particular mouse had never before been presented with a blue marble in an experimental setting. One of its peers had, yesterday, been trained to place a blue marble in a black ring in exchange for a reward. (Mice are red-green colorblind, but they can distinguish yellow and blue, and can tell light from dark.) After demonstrating its command of the task, that mouse had been placed back in the colony. Cameras had captured its interactions with the other mice and now, twenty-four hours later, I had snagged this particular mouse—one with which the first test subject had interacted repeatedly—and presented it with a blue marble and the standard array of five colored rings to which they were all accustomed.

Mice don’t talk. They are extremely social; they groom, they wrestle, they snuggle . . . they even sing to one another, after a fashion, in voices too high for people to hear. But unmodified mice have never been shown to teach this sort of complex learned behavior to their offspring or peers. Bees, on the other hand, do something similar all the time, despite their minuscule brains. Could this gene modification, just a few molecules affecting the temporal lobe, result in a significant change in social behavior?

The mouse’s whiskers twitched. Then, with a flick of its tail, it propelled itself forward and plunked the marble into the ring like a champion basketball player.

The white ring.

No treat for you, mousie.

I repeated the trial four more times, as the experimental protocol dictated, and the mouse got the black ring once. No better than chance. “Sorry,” I muttered as I gently retrieved it from the experimental enclosure and returned it to the colony, “you have to repeat a grade.”

I checked that the experiment had been properly video-recorded, logged the results, then started looking for my next subject. White mice all look alike, but the cameras in the colony could at least read their ear codes and point out the candidates—mice that had repeatedly interacted with the first test subject—on the screen for me. But I still had to catch them by hand.

Have I mentioned that mice function quite well in zero gravity?

---

I was sipping a bulb of coffee in the break room when Samantha Clarke, my friendly local neighborhood maintenance tech, drifted in. “Hey, Sam,” I said.

“Hey, Chelle.” (Only my mother calls me Michelle, and Sam and I were familiar enough that she didn’t have to call me Dr. Yan.) She sniffed theatrically. “Not too bad today.”

I gave her a middle-finger salute. She maintains that I always smell of mouse pee. I . . . wish I could disagree. I keep as clean as I can, considering my department’s limited water budget. Most of which goes to the mice.

Lasky Station is thirty-five years old. That’s, like, two hundred and fifty in people years. It was sent up by the Portuguese during the Second Space Rush, and when it was retired after many years of service it was purchased and donated to Grinnell University by billionaire alumnus Leslie Lasky. That’s why it’s named after him. Unfortunately, he cheaped out on the operating endowment, leaving us the underdogs in the never-ending battle against the forces of entropy.

Still, we’re lucky to have a station of our own. Not many small universities do. Among other things, it means that we can still do genetic research on animals, which is prohibited in most places on Earth’s surface because of the risk of modified test subjects escaping.

Which is not to say the mice never get out. They do. They just can’t escape to anywhere they could crossbreed with wild types and possibly cause all kinds of havoc.

(Unless one escapes the lab and manages to make it to a shuttle and then escapes the shuttle facility after landing. Which has never happened to me, thank god, but it happened to a friend of mine at another institution. Would have lost their certification forever if anyone had found out.)

“Catch any elves lately?” I asked.

“Nah. And I’ve tried everything for bait. Peanut butter, chocolate, even toothpaste.”

“Toothpaste?”

She shrugged. “I figure even elves like to freshen up after a snack.”

Sam and I have been ribbing each other all term about the “elves,” who she claims live in the walls and fix things if you leave out little treats for them. I don’t think she actually believes in them, despite her unwavering nominal support of the idea, but, as she says, there must be something keeping this ramshackle old station from disintegrating into a cloud of free-floating parts. It certainly isn’t stable funding and scrupulous maintenance.

“We wouldn’t have these problems,” Sam continued, “if you could just keep your test subjects from going walkabout.”

Sam has, of course, frequently proposed that the elves—for which she has never presented the slightest bit of hard evidence—might be escaped, superintelligent mice from my lab. “Yes, they are modified,” I said, just like I do every time something like this comes up, “but they are still just mice. They have literally one-tenth of one percent the number of neurons that we do.”

“Spiders have even fewer neurons,” Sam replied, “and yet they are mighty hunters. Almost as good as cats.”

“Where’d you read that?” I countered, surprised.

“Haven’t you kept up with your field, doctor?”

“Can’t read everything,” I muttered, making a note to look that up later.

---

In the afternoon I did maintenance work on the colony. The mice lived better than I did, frankly. They were provided with food, water, clean bedding, spaces for both privacy and socialization, and lots of enrichment activities, and they didn’t even need to do their own laundry. They were free to go wherever they wished within the colony and do whatever they wanted, and the population was carefully controlled so they were never overcrowded or isolated. And there were plenty of colored marbles and rings scattered about so they could practice what they’d learned in the experimental enclosure . . . and hopefully share it with their friends. Apart from the complete lack of privacy and the inability to leave, it was basically Mouse Utopia. “You wouldn’t want to go outside anyway,” I told one mouse as it made a break for the door. “It’s very, very cold out there.”

I always counted the mice in the main space before I exited the colony, to make sure none of them slipped out the double door with me. They were fast and adroit, and in the early days occasionally one would make it past the first door. But I’d learned their tricks and it had been a long time since one got even that far.

Yet still they went missing. The cameras were omnipresent and their census was automated and very thorough; there was no doubt that eighteen mice had somehow managed to vanish from their enclosure in this term alone. I even knew exactly which mice they were, by their ear tags. But though I had studied the recordings and even tracked the individual mice right up to the hour of their disappearance, I hadn’t been able to figure out how they were escaping.

This was tremendously annoying to me.

The whole point of running a mouse lab in space—the reason we put up with the dehydrated food and the constant noise and the loss of muscle and bone mass and, oh yeah, the ever-present threat of sudden death from equipment failure—is to keep the mice from getting away. And, yes, I could be fairly confident they hadn’t gotten back to Earth, which was the primary rationale . . . and there was the view and the zero-G sex, which are both as great as you’ve heard. But it was still worrisome that, despite everything, I’d managed to lose track of eighteen of the little buggers.

---

“It’s the smell that bothers me,” I complained to my department chair, Dr. Prentis, on our weekly call.

“You should be used to it by now,” she said. Her face was streaked with glitches and there was a noticeable lag. We must be on the other side of the world from Iowa at the moment.

“Sorry, I meant the lack of smell. If they were getting out of the colony and dying in the walls somewhere, we’d smell the decomposition. Or, if they actually found a way to survive, we’d smell their leavings eventually. But no one has mentioned it. And, believe me, if anyone on this station smelled anything even vaguely mouselike outside of my lab I would never hear the end of it.” I blew out a frustrated breath through my nose. “They aren’t in the colony. So where are they?”

“Here’s my best guess. They are making it to an airlock and getting blown out the next time it opens.”

I nodded. There were numerous small airlocks for service, maintenance, and disposal, as well as the main cargo and personnel locks, and they might seem an attractive place for an escaped mouse to settle down. For a while. “Possible. But all of them?”

She shrugged. “As long as they aren’t damaging the station or escaping to Earth, and as long as you record the losses in your logs, I consider this a low priority. Now let’s talk about those error bars.”

---

Another day, another experiment. This was run one hundred twenty-eight, phase A, which meant that I was training a mouse to put a marble in a ring. This part of the experiment was more immediately satisfying, because the mice and I both knew how to do it. It was just a matter of repeating the behavior and the reward until the mouse could perform the behavior reliably. But it was also boring, and Not Science, because the results were so well understood. Phase B—the part that so often didn’t work the way we hoped it would—was more interesting, and Definitely Science, but also the most frustrating. It wouldn’t be until I’d collected all the data and returned to Earth with it that we would know whether or not we had done something worth publishing.

It wasn’t looking great so far. The phase B mice seemed to be doing a bit better than chance, but we wouldn’t know until we’d crunched all the numbers, controlling for things like age and sex and the presence of certain other genes, whether the difference was statistically significant. If it wasn’t significant, that implied that our whole theory about the function of the superior temporal sulcus in social communication and learning was a bust, at least in the mouse model. Of course, a negative result is still a result . . . but they don’t give Nobel Prizes for negative results.

One thing I had noticed—anecdotal evidence, I suppose, though can something be anecdotal if you haven’t told anyone else about it?—was that certain phase A mice seemed to be much better teachers than others. Or just lucky. At this point it was impossible to be sure.

“Tell me about your friends,” I said to the mouse I’d just caught as I carried it to the experimental enclosure. “Are you close? Do you like to do things together? Will they be excited to hear about what you learned on your vacation? Will there be slides?” I placed the mouse inside and closed the door. It quickly oriented itself to its new environment and looked back at me through the scratched, transparent plastic. Its nose wiggled adorably. “I hope you are an engaging speaker. And for god’s sake don’t read every word on every slide.”

We watched each other for a while, the mouse’s red eyes meeting mine unblinkingly. What was it feeling? Terrified of this giant, incomprehensible creature? Eagerly looking forward to a new adventure, and maybe a treat? Bored? Horny? Planning escape?

Was it wondering the same things about me?

The computer had randomly selected a marble color and ring color, and all the mice had been exposed to the experimental setup before, so it wasn’t long before the mouse found the marble and put it into one of the colored rings. Wrong ring, mousie . . . no donut. A soft buzzer sounded and the marble vanished into the mechanism, replaced by another one of the same color.

The mouse wasn’t discouraged; it kept trying, and it wasn’t long before it happened to hit upon the correct ring. A light flashed, a beeper beeped, and a pellet emerged from the dispenser, which the mouse immediately pounced upon and devoured. The mechanism then produced a marble of a different color. This part wasn’t as much fun . . . no matter which ring the mouse selected, the marble would disappear with no sound at all and certainly no treat. Life isn’t fair.

We played this thrilling game for an hour, with me carefully selecting the color of the next marble to optimally reinforce the desired behavior and the mouse catching on quickly. Good for you, mousie. When we reached the point that, when presented with two marbles of different colors, the mouse selected the correct marble and placed it in the correct ring five trials in a row, I declared victory and pulled out. I delicately scratched the champion’s ears and gave it a little smooch on the head—though with my mask and face shield, it was more of an air-kiss—before sending it back into the colony to preach the gospel to all creation.

And then I heard a bang.

That wasn’t good.

Everyone knows that there is no sound in space. Everyone is wrong; space is loud. At least, the inside of a space station is. Fans whir, plumbing gurgles, structures creak, a thousand electronic things bleep and burble. Almost everyone has difficulty sleeping. But a bang, especially a sharp metallic bang like this one, is very uncommon and is never good news. Could be an explosion; could be an overstressed structural member giving way; could be a meteoroid strike.

A moment later a klaxon began sounding. Air pressure drop.

That really wasn’t good.

---

“Okay, here’s what we know.”

We were gathered in the main dining hall in module C, all sixty-three of us, peering past each other’s shoulders, many in the corridors outside straining to hear. Dr. Ivanovic, the station administrator, looked even more worried than usual.

“We’ve taken a micrometeoroid strike,” she said. “It’s somewhere in module A. But we don’t know where.”

A sound combining a groan and a sigh came from the assembled crowd. A groan because it was bad news, and a sigh because this was typical for Lasky Station.

Lasky was an old-fashioned design, with module A—the oldest and largest of the station’s eight modules—as the central spine to which all of the others attached. It couldn’t be subdivided, and if it lost pressure, the station as a whole would be unusable. There had been talk for years of upgrading the station to a more modern, robust design, but that would require money we didn’t have.

This is what happens when a small liberal-arts school has its own space program, I guess.

“We are losing atmosphere, albeit slowly.” More groans. She patted the air with her hands, signaling for calm. “We have people working the issue. Inside, there’s a lot of equipment that has to be moved. Outside, drones have failed to spot the impact point as of yet.” The outside of the station, I knew, was just as much of a mess of accreted junk as the inside. Even worse, perhaps, because obsolete and unused equipment on the station’s outer hull wasn’t in anyone’s way, so it tended to just stay where it was. And then something else got bolted to it, so now there was even more reason to leave it in place. Et cetera ad infinitum.

“So what’s the ETA for a fix?” came a voice from the back of the crowd.

Dr. Ivanovic sighed and shook her head. “Don’t have one yet. But if we don’t manage to find the leak in . . . ?” She shot an inquisitive glance off to the side. Her face showed she didn’t like what she got in response. “ . . . twenty-one hours, we’ll have to begin evacuation procedures.”

The crowd got really ugly then, with people shouting “I can’t possibly . . . !” and “What do you mean . . . !” and “What kind of . . . !” and no amount of air-patting would bring it under control. I was one of the ones shouting, I must admit.

Dr. Kwok, head of Facilities, finally screamed loud enough to be heard over the pandemonium. “Stop behaving like academics!” she yelled. “This isn’t about who took the last donut in the break room! This is life and death!”

That shut us up.

“Twenty-one hours,” Dr. Ivanovic repeated into the embarrassed silence. “If anyone from the repair crew asks for your help, make that your top priority. Otherwise . . . use that time to get ready to evacuate.” As we muttered and jostled our way out of the lab, she called out, “And remember: two kilos each, max. No exceptions. Anything that can be replaced, no matter how expensive—leave it here.”

I stopped dead in the corridor, not really hearing the annoyed exclamations of the people bunching up behind me.

My mice. My unique, purebred, genetically modified, carefully trained mice.

Shit.

---

I wasted an entire precious hour arguing with the administration, both locally and back in Iowa, but there was no possibility of a reprieve. There were only two shuttles, and they had only so much capacity. Not to mention that my mice were illegal on Earth, which I’d managed to forget in the initial moment of panic.

So if we had to evacuate—and by the end of the hour nothing had changed on that score—the mice would have to stay behind. All of them. And even if I didn’t euthanize them before leaving, which I wasn’t yet ready to consider, given that we were evacuating because of an air leak, they would likely die before the station could be repaired and reinhabited. Assuming we didn’t just decommission it, which given our finances was the more likely outcome.

I floated in the middle of my lab, face in my hands, gently bumping into walls, trying to control my breathing. Filled with anger and regret and . . . and grief. Grief for the project, certainly, for the years of work that would be lost, but also grief for the mice. The innocent lives that would be snuffed out.

Yes, okay, I had culled a lot of mice in my time. Every pup that wasn’t genetically correct, every juvenile that fought with its peers, every adult that got sick or injured in a way that couldn’t be easily corrected. Gently, painlessly, respectfully, but still culled. Killed. But most of my mice lived out full lives, well cared for and, I hope, provided with a pleasant and interesting environment. And their lives had purpose! They served the advancement of human knowledge. But if they all died now . . . those lives, and the lives of their predecessors, would simply be wasted. We didn’t even have enough results yet to be statistically significant, and when we resumed the experiment—if we even did—we’d have to start from scratch.

I had been given a lightweight plastic box for my two-kilo allowance. I began filling it with data bricks, tissue samples, vials of crystallized DNA. It didn’t seem like much, for all those years of work, but it filled the box surprisingly quickly.

As I worked I dislodged a yellow marble—the damn things got everywhere—and absentmindedly caught it before it could get away and cause trouble. The stupid little things were like Barbie shoes, omnipresent, always getting caught in some mechanism or clogging an air vent.

Then I stopped and looked at it.

Held it between my fingers.

Looked at the tidy array of five rings in the experimental enclosure.

And went off in search of the nearest emergency equipment locker.

---

“Okay,” I said to the mouse, “this is a micrometeoroid repair patch.”

The mouse just stared at me.

I’d picked the best mouse for the job. I hoped. Its ID number was 23070418—born on the Fourth of July, hooray—and according to a quick crunch of the experimental results it was the best of my “superteachers.” Or, as I said before, just very lucky. At this point I didn’t care which, as long as past performance was indicative of future results.

I tried not to think about the fact that that phrase usually appears in the negative.

I’d made some modifications to the experimental enclosure. I’d removed the marble dispenser, leaving an opening into the enclosure, and added to the five rings a crudely hacked hole, which was glued to a tube, which was taped to a hose, which led to a vacuum cleaner in the next room. (Yes we use vacuum cleaners in space, and yes I have heard all the jokes.) The cleaner was not yet running. That would come later.

What I was about to ask this mouse to do was certainly outside its experience and very likely beyond its capacity. But I had to try.

I pushed the patch through the hole. The mouse first skittered away, then approached it hesitantly. I had lightly smeared the patch with peanut butter, to make it attractive and to mask the smell of the chemicals within.

At first the mouse simply sniffed at the patch—a two-centimeter circle of tough fabric with a bulge in the center. Then it licked at the peanut oil.

I held my breath.

It delicately nibbled at the patch.

Good mousie! I pushed the reward button. Beep, flash, treat. The mouse abandoned the patch and went off for the treat, as I’d known it would, but it was a start. I retrieved the patch and prepared another one.

You’ve probably seen videos of squirrels running through amazing obstacle courses of anti-squirrel measures in order to get food from a bird feeder. This generally happens because the bird feeder owner, deliberately or accidentally, set up the obstacles one by one as the squirrel learned to overcome them. And that was what I was trying to do right now: to inculcate a series of arbitrary tasks in order to achieve a reward. Except the tasks were not arbitrary and the ultimate reward was not a treat, it was life.

I would most likely only have time to do this once. At most. I had to hope that would be enough.

I ran through an entire box of patches and had to steal another one from another equipment locker. What I was doing was a violation of safety standards and several codes of honor and ethics, but lives were at stake.

Okay, they were only mouse lives. But they were lives nonetheless, and my career and the careers of my colleagues back in Iowa were also in jeopardy. I kept working, ignoring the noise of everyone else running around getting ready to evacuate and listening to the increasingly anxious announcements on the public address system with only half an ear.

I did hope that my work would not be needed—that some human being would find the hole and patch it before it was too late. But what I was hearing made that sound increasingly unlikely.

---

I reached into the box for another patch, and cursed as my fingers found it empty. But I felt I was getting close; I went and found another box. On my way back, I ran into Sam.

“What are you doing with those patches?” she asked. “Did you find the leak?”

“Not yet,” I temporized, trying to push past her.

“But if you even think you’ve narrowed it down—you have to let Dr. Kwok know right away!” She kept blocking my way, and she was better at free-fall maneuvering than I was.

“I’m . . . I’m just working a hunch, okay?”

“But we might need those patches!”

Then I saw an opening and leaped, pushing off the wall behind me and shooting past Sam. “I’m going to give them to the elves!” I called as I passed.

I hoped that she’d be too busy with evacuation prep to follow me. And, indeed, she didn’t. But this encounter reminded me that my time was extremely limited.

---

I had finished up the third box of patches and had just started on a fourth when we had a breakthrough. The mouse caught the patch as soon as I pushed it through the hole, then began nibbling at the edge of the paper covering the adhesive. This wasn’t an easy task, but with teeth and busy little paws it managed to pull the paper away, exposing the adhesive, without getting its nose or whiskers stuck. Then, with a neat little push of its back legs, it propelled itself toward the hissing hole where the tube from the vacuum cleaner—now running—was attached.

Now came the part that it had failed in the last few tries. I held my breath.

For a moment the mouse, clearly hesitant of the noise and moving air, held back. Then it moved toward the hole, holding out the patch . . . the wrong way up. Again. I let out my held breath as a sigh and began preparing the next patch.

But at the last moment the mouse turned the patch over and released it. The vacuum pulled the patch onto the hole, sticky side down, forming a tidy seal.

Then the mouse, I swear, turned and looked me right in the eye.

I looked back at it, stunned.

I mashed the reward button. Beep, flash, treat.

I reset the experiment as the mouse was enjoying its reward—the adhesive took a little time to set, fortunately—and tried it again. Okay, mousie, you wowed ’em at the matinee. But are you ready for the main event?

It was. It nailed the task. And a third time, and a fourth, and a fifth.

I hammered the reward button. A cascade of beeps, flashes, and treats gushed into the enclosure, which the mouse immediately snagged from the air and devoured.

Congratulations, mousie, you’re a star!

I retrieved the star from the experimental enclosure, giving it a gentle scritch behind the ears—its eyes closed in pleasure—and released it into the colony, along with the remaining contents of the fourth box of patches. “I’m sorry the career of a prima ballerina is so short,” I told it as I closed the door. “But now you retire to what I deeply hope is a successful second career as a teacher.”

I took in a breath, let it out. It’s all up to you now, mousie.

I checked my watch. Thirteen hours left. Not as much as I’d hoped, more than I’d feared. More than I’d had any reason to expect, frankly—this mouse was a freaking genius. I had to hope it was a genius teacher as well, because I could only allow it . . . let’s say six or seven hours to pass along what it had learned, before I started moving the students into module A. I figured I could smuggle at least a dozen of them at a time in a carrier under my lab coat.

And then there came a knock on the door.

---

“Dr. Ivanovic!” I cried. “What a . . . what a pleasure.” This was, I was pretty sure, the first time the director of the whole station had come to visit my lab. And alone, to boot. She entered without a word and closed the door behind herself. She seemed troubled—not meeting my eyes—and I was sure she was about to give me some bad news.

“I was a mouse handler myself, in my grad student days,” she said by way of introduction. This wasn’t something I had known. “So I understand the . . . concern you must feel for your charges.”

I nodded, not trusting myself to say anything. Sam must have gone straight to the top with the news that I was stealing repair patches. I was in serious trouble.

Now she did meet my eyes. “Now you have to promise to keep this news to yourself.”

“I promise,” I said, my guts in an uproar. Bad news for sure, but maybe not the news I’d been expecting?

“There’s been no progress in finding the leak. I’ve been told that the chances of avoiding evacuation are extremely slim.” She reached out for my hands, and I numbly accepted her grasp. “It is time to begin the . . . the unfortunate but necessary task of euthanizing your animals.”

My eyes darted involuntarily to the colony. There was no sign of my genius teacher. I hoped it was already inside, conducting the introductory lecture. “I . . . I can’t do that.”

Dr. Ivanovic squeezed my hands, gave me a motherly look. “I understand how you must be feeling, Dr. Yan. But it’s better that they die by your hand—a quiet, respectful death—than to suffocate as the air runs out.”

“I hope we can avoid both of those.” But I couldn’t possibly tell her why I had that hope. Frankly, it was an insane plan—barely deserving of the name “plan” at all—and almost certain to fail. But if I had to euthanize the mice . . . there went even that tiny sliver of hope.

She squeezed my hands again. “There is a time for hope, Dr. Yan, and a time for realism. We are scientists. We understand that facts are facts. That means accepting that sometimes a theory doesn’t bear out . . . sometimes an experiment fails. And when that happens, we cut our losses and move on to the next thing.”

I could have given this speech to one of my own undergrads. In fact, I think I did, just last term. But that didn’t make it any easier to hear. I pulled my hands free of hers. “I . . . I need some time to prepare.”

She sighed, as though disappointed. “All right. But don’t take too long about it, okay?” She turned to leave, then paused. “Would you rather have someone else do it? I can send one of my grad students.”

“No! I, I mean, ah, no thank you. I . . . I need to do this myself.”

“I understand. I’ll check back in a bit and see how you’re doing.”

Shit. “Thank you so much.”

“You’re welcome. This is a hard time for all of us.”

---

I closed the door behind her. It was just a partition door, not an airtight hatch. It didn’t even have a lock. And barricading a door in free fall? Not really a thing. I couldn’t exactly push a heavy cabinet up against it.

I needed to give the mice time. Time to learn, time to make their way to module A, time to find and fix the hole.

I thought about welding the door shut. I had a soldering iron, but no welding torch. I thought about putting up a biohazard sign. “Mouse plague!” I’d moan weakly. “Save yourselves!” But they wouldn’t believe me. Or else they would believe me and leave me behind, which given my own assessment of the plan’s chance of success didn’t seem like a great outcome either.

This was a matter of survival. And there’s nothing better at surviving than mice. They’ve made it through millions of years of evolution despite the fact that everything wants to eat them.

So. Think like a mouse.

There’s a lot of random stuff in a bio lab. I set to work.

---

It was about an hour and a half later that my pocket comm trilled. It was Dr. Ivanovic, of course. I ignored it and kept working. A few more calls and then came a text message: “OPEN THE DOOR.”

“Open it yourself,” I texted back, and kept working.

I was surprised that it took nearly another hour before the crash of the outer lab door opening came to my ears. I guess she thought I might come to my senses if she waited? Or maybe the glue I’d used had been more of an impediment than I’d expected.

After the crash came a tearing sound, followed by cursing. A few minutes later another tearing sound, then more vehement cursing. And again. Then a pause. “This isn’t funny!” Dr. Ivanovic yelled. I could barely hear her, and didn’t bother trying to reply.

It wasn’t intended to be funny. It was intended to be annoying and time-consuming. I’d glued ten sheets of tough plastic across the width of the outer lab at half-meter intervals. Even with a sharp knife it would take her some time to cut through them all.

I kept working.

Given time a mouse, or a squirrel, will overcome any number of obstacles to obtain a reward. But if you present those obstacles one at a time, the subject will tackle them one at a time rather than stopping, thinking, and finding another way around. They are, in some ways, no smarter than administrators.

The tearing and cursing stopped for a while. Then I heard new voices, and a buzzing whir. Sounded like she’d come back with help, and they’d brought a hot cutter, or maybe something even more destructive.

I’d anticipated this. But they made much quicker progress than I’d hoped, the sounds getting louder and louder as one plastic sheet after another gave way. I checked my watch and redoubled my efforts. I’d bought the mice nearly three hours already. Four, I thought, was the minimum; five would be better; every minute after that would be gravy.

The buzzing whir got louder still, then stopped. “What the actual fuck?” came an unfamiliar voice.

I smiled.

The space between the last plastic sheet and the inner lab door was filled with polymeric polyol—a tough, sticky foam resulting from a reaction between two common organic chemicals. It was a trick I’d learned as an undergrad—I’d been the victim of a nasty chem lab prank, and had learned the formula in revenge—and it was messy, stinky stuff, which stuck to skin and had a tendency to clog up anything you used to try to cut it.

After a period of muttering and shouted threats, accompanied by a fusillade of ignored signals on my comm, the buzzing whir started up again. A moment later there came a zownt sound and a gurgle, and a god-awful stink I could even smell in here. It must be ten times worse out there. The sound of the cutter stopped, replaced by rapidly diminishing coughing and gagging.

That tube full of butyric acid was just one of the surprises I’d embedded in the foam as an enrichment activity for my adversaries. It’s important to keep your test subjects’ minds active and engaged.

I kept working.

---

By the time they made it through the foam and I could hear them hammering at the lab door, which I had simply wedged shut, I had nearly completed my to-do list.

I had cut numerous holes in the walls, leading to the air ducts, the spaces between walls, and the space between the inner and outer hull. Check.

I had dumped two boxes of repair patches into those spaces, using a blower to move them as far in as possible. Check.

And I had gently carried the mice, in groups of three or four, out of their colony and released them into those holes.

Every single mouse.

Check!

I felt accomplished. I’d managed to save all the mice from euthanasia—for the moment, at least—and stolen between four and a half and five and a half hours for their training. Using the cameras in the colony, I’d snagged and released the ones who had had the most and earliest contact with my superteacher first, leaving the less enthusiastic students in place for makeup classes. But now, with—I checked my watch—six hours left before we were to begin evacuating, they had all at least had an opportunity to be trained, and now were free, with what I sincerely hoped would be sufficient time and resources for at least some of them to make their way to module A, find the leak, and fix it, saving their own lives in the process.

Personally, I was completely fucked. But I’d done what I could for the mice, and for the station.

The door crashed inward and two burly grad students in gas masks came tumbling through. They seemed surprised that no further obstacles lay before them. Just me, floating placidly in my badly deranged lab, holding up my hands.

“Uh, Dr. Yan,” said one. “We, uh, I’m sorry, but Dr. Ivanovic sent us to euthanize your mice.”

“What mice?” I asked innocently.

---

We don’t exactly have security staff on Lasky Station, but we do have Dr. Kwok, and after some searching she came up with a pair of cable ties, which she used to fasten my wrists to a structural member in one of the escape shuttles. She wasn’t gentle about it.

“You are in deep shit,” she reassured me, “but I don’t have time to deal with you now. You just stay here, stay out of the way, and try not to get into any more trouble. We’ll deal with you when we get back to Iowa.”

She didn’t even leave me my comm, the bitch. But my hearing is pretty good, and I was able to follow the ongoing story from what echoed through the station and into the shuttle. And occasionally someone would come into the shuttle to fetch or carry something; a few of them were even still speaking to me.

For the first couple of hours most of the staff were still focusing on getting themselves and their labs packed up for evacuation, while the Facilities crew continued, with ever-diminishing confidence, to try and find the leak. But gradually the emphasis changed, with more and more announcements asking people to please finish up packing and help deal with the mice. Over time the evacuation morphed into a full-on mouse hunt. And the mice were clearly winning. I was ashamed of myself for causing so much trouble, and at the same time smugly proud of my charges.

I would say I was also concerned for my career, but I knew better; that ship had sunk. I would have to find some other way to pay back my student loans. If I were lucky, I wouldn’t be doing it from prison.

Suddenly everything changed. There was a loud and rather frenetic announcement on the public address system, which I couldn’t quite make out, but it was followed by raucous cheering, which then devolved into scattered applause and a babble of conversation. I didn’t know for sure what had just happened, but I had my suspicions . . . and my hopes.

They left me stewing in the shuttle for another hour before Dr. Kwok came and got me. “Can you explain . . . this?” she said, thrusting something in my face.

It was a mouse.

It was alive.

It had a micrometeoroid repair patch stuck to one front paw.

And it was not my genius superteacher.

Yes, lab mice all look the same. But a teacher, even a postdoc, doesn’t forget her best student.

---

It was touch and go for a while there, but they let me stay on the station. Just long enough to recapture the mice, at first, but eventually everyone admitted that we’d stumbled onto a significant new result in the genetic basis of socialization and learning—and teaching, which we hadn’t expected at all—and there wasn’t anyone better than me to follow through on it. Or any better facility or population of test subjects.

I’m now a principal investigator, with a staff of three grad students. The mouse lab has expanded to completely fill module D, and the Lasky Foundation was persuaded by the publicity to increase our operating endowment. We are still underfunded, of course, but less so than before.

Mouse number 23070418, now called Maria Montessori, retired from the ballet completely and became a mascot, living in a small luxurious cage in my office—yes, I have my own office now!—and gratefully accepting treats from her legions of fans.

And as for the rest . . .

Well, I must confess I never recaptured all the mice I released that day. Most of them, yes. And I found thirteen little corpses, eventually. But that still leaves thirty-one unaccounted for, not including the original eighteen. And there’s still no smell of mice, alive or dead, anywhere on the station outside the mouse lab. It is a puzzlement.

And people keep coming to me with strange stories of missing items mysteriously reappearing, or small problems being repaired, if you just leave out treats overnight.

They want to know what the best kind of treats might be.

“How are the elves today?” Sam asks whenever we meet up in the break room.

“What elves?” I always say. And wink.