Eleven Milligrams
The tray came out of the nitrogen cabinet with a paper slip taped to the lid, and the slip of paper had arithmetic on it. Sanae Ozaki had read the slip four times that morning before she said anything, which is how many times you read a parking ticket before you believe the city really meant it.
Site 1. Trays A through F. Catalogued particles, twenty microns and up: 3,411. Fines below twenty microns: 4.17 grams. Site 2: 2,877 particles, 2.98 grams of fines, the entire site sealed as reserve end-member since intake. Ten grams and change of a moon nobody had ever stood on, catalogued grain by grain, sitting in a cabinet in Sagamihara with her request form clipped to the outside of it.
Her request form had a box on it that said MASS REQUESTED. She had written 400 mg in the box. The response, in Nao Miyake’s handwriting, was the number 11.
“Eleven milligrams,” Sanae said.
“Eleven milligrams,” Nao said. She was standing on the other side of the pass-through with her arms folded, and she had the look of a woman who had already had this conversation with the paperwork and won. “Site 1 fines. Four replicates at two and a half milligrams, one milligram held back for blanks and transfer loss. I put your name on the allocation sheet this morning.”
“Four replicates.”
“Which is a population.”
“It’s a start on a population.”
“It’s four independent measurements of the same homogenized fines. At your precision that’s a Δ17O with two zeros after the decimal point and a standard error small enough to walk through a wall. You want the origin question, that number answers the origin question.”
Sanae looked at the tray. Ten grams and change, and they were haggling over the mass of an eyelash. The capsule had come down at Woomera on a Thursday in September with a parachute the color of a traffic cone, and eleven days later this was where it had ended up: a nitrogen cabinet, a curator, and a paper slip.
“I want four hundred,” she said, “because my protocol calls for twenty milligrams a replicate, ten replicates per site, both sites, with density separates. I want the mean and the variance, Nao. I want to know if this body is one thing or two things ground together.”
“Then you want Site 2, and Site 2 is the reserve.”
“The reserve is for methods that haven’t been invented.”
“The reserve is for a person who hasn’t been born. Same thing, longer fuse.”
Nao had been the curator of this collection for nineteen years before the capsule landed, back when the collection was a shelf of lunar dust and a few grains of Itokawa that the facility held on loan. Her signature was the only one on the chain of custody for the Phobos material. Not a counter signature, not a witness line. One name, thirty years deep, on every transfer form since intake. Sanae had looked this up once, late, after the Tucson preprint came out, and had spent an unwise amount of time staring at a PDF of the transfer log thinking about what it meant that a whole moon’s worth of material could pass through a single person’s handwriting.
“I’ll put the Site 2 release on the November board’s agenda,” Nao said. “With a witness present.”
“The review board of my institute sits in October. Yours sits in November. Mine folds my group into Tanaka’s lab on the twenty-second unless I walk in with a number.”
“Then bring them the eleven milligrams.”
“Eleven milligrams of Site 1 fines is the mean of the most reworked ground on the body. Stickney’s ejecta blanket. If Phobos is a captured asteroid, Site 1 is that asteroid after four billion years of being hit. If Phobos is Mars’s own crust thrown out by an impact, Site 1 is that crust after four billion years of being hit. Either way the mean tells you what the gardening produced. I need the populations, Nao. I need to see the distribution.”
Nao looked at her for a moment. “Then request the grains.”
“I did.”
“You requested four hundred milligrams.”
“I requested what my protocol calls for. The grains are in the same box.”
“And the board will read your form the same way I did.”
The imaging argument happened two days later, on a Tuesday, in the isotope lab two doors down from the cabinet, because Sanae had brought Tomás Iriarte’s latest mosaic and she wanted Nao to see it on a big screen instead of a phone.
The mosaic was the whole of Site 1, stitched from the curation microscope: thousands of grains laid out on the antistatic sheet in their catalog order, gray on gray, every one numbered. Tomás had color-coded them by morphology. Most of them were dark, angular, fine-grained, the color of wet asphalt. A few dozen were pale. Three were nearly white.
“Non-destructive first,” Sanae said. “SEM, micro-CT, reflectance on every catalogued particle. Tomás can run it in six weeks and it costs you nothing but instrument time. Then we pick targets. Then we spend the targets.”
“Targets for what?”
“NanoSIMS. One grain, one number. Spatial precision at a sputter.”
Nao looked at the mosaic. “You know what imaging doesn’t give you.”
“Oxygen isotopes.”
“Oxygen isotopes. You can image a grain until you have a doctoral thesis about its shape and you still haven’t measured a single atom of anything. Spinel looks like spinel. A CAI looks like a white speck. What you want is a mass spectrometer, and a mass spectrometer eats the sample.”
“NanoSIMS leaves a crater.”
“A crater in a twenty-micron grain is most of a twenty-micron grain. Don’t call it a crater. Call it what it is.”
Sanae let this go, which was the point of having brought the mosaic. “Can I make the case for the sputter, then.”
“Make it.”
She had made this case to her group so many times that Rin could deliver it in her sleep, and she made it again anyway, to the curator, standing in front of the picture of ten thousand grains.
“Mars has its own line.” She put her finger on the empty space in the air where the plot lived, the one she had been drawing on whiteboards since she was Tomás’s age. “You plot oxygen-17 against oxygen-18, and terrestrial rocks fall on a line, and everything in the solar system that has been through a fractionation process falls parallel to it or off it. Mars sits on its own line, three tenths of a per mil above ours, because Mars has a mantle and Mars’s mantle has a mass-dependent fractionation history and nobody else’s mantle has ever shared it. Carbonaceous material sits below the terrestrial line. CI, CM, all the primitive soup, minus two tenths, minus three tenths. There is no third place to stand.”
“I’ve read the same papers you have.”
“Then you know the separation between the two hypotheses is six tenths of a per mil, which is a canyon. Any instrument on this floor can throw a rock across it. The problem isn’t the precision, Nao. The problem is the word mean.”
“Explain it to me slowly.”
“If Phobos is captured, it’s carbonaceous all the way down, minus three tenths, and every grain you average will sit below the line. If Phobos is Mars ejecta, it’s Martian all the way up, plus three tenths, and every grain will sit above. Those are the two clean cases and a bulk measurement of eleven milligrams will nail either one and I will happily retire on it.” She paused. “What if it’s neither.”
“Go on.”
“What if it’s a Mars ejecta body that has been collecting carbonaceous impactors for four billion years, or a carbonaceous body that has been collecting Mars ejecta for four billion years? Both are real. Both land material on the surface. Phobos sits in Mars’s equatorial plane at three radii, it’s tidally locked, and the Mars-facing hemisphere takes the Martian flux like a face in the rain. Average a mixed regolith and you get a number between the two lines, and the number between the two lines is a number that supports whichever paper cites it. Tucson will put your intermediate mean in their introduction and nobody will ever resolve it.”
“And the crater in the twenty-micron grain resolves it.”
“One grain is not a population, I know what I said. But a distribution of grains resolves it. Twenty grains is a scatter plot. Fifty grains is two clusters or one. That’s what NanoSIMS buys you that fluorination can’t: not precision, deconvolution. A bulk charge of two and a half milligrams is ten thousand grains averaged into a single number before the oxygen ever reaches the mass spectrometer. You can’t un-average them afterward.”
Nao was quiet for a moment. Behind them, the fluorination line sat cold along the far wall, its brass manifold wrapped in heating tape the color of old bandages, its vacuum pump ticking as it held a vacuum against nothing in particular. The line was rated for three milligrams a charge. Sanae had used it for eight years. She knew how the pressure rose when the BrF5 went in, the little jump and settle, like a throat swallowing.
“And the CAI,” Nao said.
“One CAI-type spinel, magnesium-aluminum oxide, condensed out of the solar nebula at seventeen hundred degrees before Mars was a rock, and it sits twenty per mil below anything the Martian mantle has ever produced. Twenty. Not three tenths. Twenty. One grain, one session, and the giant-impact hypothesis is a story people tell at conferences while their sample committee schedules another meeting.” Sanae tapped the pale specks in Tomás’s mosaic. “Tomás thinks three of these are candidates on morphology alone.”
“On morphology alone.”
“On morphology alone, which is why I’m asking for imaging first and NanoSIMS second and not the other way around.”
“Which is also why your three candidates might all be plagioclase, and you will have spent three catalogued grains of a sovereign moon to learn that they are plagioclase.”
“One of them won’t be.”
“And if it is a real CAI, and it is a contaminant from my own laboratory, then I have spent a catalogued grain to prove that my laboratory has a hole in it, and the joint review board will lock this cabinet for a year while they audit my nitrogen.” Nao’s voice had not risen. It never did. “That is the counter-move, Sanae. Not the physics. The chain of custody. A single exotic grain proves its own anomaly and nothing else, because a population of one has no error bar, only a story. You taught me that sentence. I’m giving it back.”
Sanae looked at the mosaic for a while. The pale specks stared back, three grains out of three thousand four hundred and eleven, and they might have been CAIs and they might have been feldspar and they could not tell her either way, because light is not a mass spectrometer.
“Fine,” she said. “You’re right.”
“Thank you.”
“Put the Site 2 release on the November board. Put my grain request on the same agenda. I’ll take twenty catalogued grains per site and I’ll take witness signatures and I’ll take the nitrogen audit, and I’ll present it all to you in November with error bars like a person.”
“I’ll put it on the agenda.”
“My institute sits on the twenty-second.”
“Then bring them the eleven milligrams.”
Sanae gathered up Tomás’s mosaic file and her notebook and her coffee, which had gone cold hours ago, and she was at the door when Nao said, without turning around, “The Americans get one percent of anything over a gram. By protocol. It’s on the intake agreement, page four, paragraph two. Halvorsen’s group could request sixty-one milligrams of those fines tomorrow morning and I would have to fill the order.”
“I know.”
“He hasn’t requested it.”
“He doesn’t need to,” Sanae said. “He already has the answer.”
“Then maybe he’s smarter than both of us.”
“He has a preprint with forty-seven citations in six weeks, and the only measurement in it is a spectrum.”
“The citations don’t know that.”
The four of them had noodles in the break room at eleven at night on the fourteenth, three days before the packet was due, and Miriam Osei did the arithmetic on a napkin because the napkin was there and because Miriam did arithmetic the way other people breathed.
“Ten grams and change,” she said. “Call it a hundred and five billion grains if you count below the micron. The whole Earth budget for this mission, and here we are fighting over a thimble’s thimble.” She wrote 11 in a circle on the napkin and underlined it twice. “Eleven milligrams. Four replicates. If I run the line at full efficiency I get your population mean by Thursday and you can put it in the packet with a straight face.”
“And if the mean is in between,” Rin Kubo said.
“Then the packet says in between, and in between is not an origin, and the panel folds us into Tanaka’s lab and I update my CV, which I have been doing every September since I was twenty-six.” Miriam said this flatly, with her mouth full, which was how she said everything. She had a daughter. The daughter was seven and was going to be eight in February and had already asked, in Miriam’s telling, whether Mama’s new office would have a window. Miriam told this story often. She told it as a joke. It was not a joke and everyone in the room knew it, and that was why she kept telling it.
Tomás, who had been quiet, said: “The grain you take Saturday.”
Sanae did not look up from her noodles. “What about it.”
“A twenty-micron olivine from Site 1 is maybe ten nanograms. You load it, you sputter it, you get a Δ17O to plus or minus a tenth and the grain is a crater. If the number comes back carbonaceous, you have contradicted yourself in public. If it comes back Martian, you’ve proved implantation or provenance, which the panel will correctly say is not an origin. So why Saturday.”
“Because the packet is due Friday and the number exists Thursday if I book the session now.”
“Under what authorization?”
The room got quieter when a graduate student asked a question she already knew the answer to.
“Mine is suspended,” Sanae said. “After the July request, the access committee flagged my account pending the board’s decision on the Site 2 matter. Rin’s isn’t.”
“Rin’s is a student certification,” Miriam said. “Student certifications are for supervised sessions.”
“Then supervise me,” Sanae said. “I’ll be in the room. I’ll sign the bench log.”
“The bench log takes the code that opened the door,” Tomás said, “and the code that opened the door will be Rin’s, and the access log does not have a field for who was standing next to the person whose code opened the door.”
“It doesn’t.”
Rin put down her chopsticks. She was twenty-four. She had been in the group fourteen months, which meant she had spent fourteen months of her adult life in the orbit of a question that was going to decide whether she had a thesis or a very interesting master’s report. “My code,” she said.
“Your code.”
“Because if I don’t, Miriam’s contract ends in June and Tomás goes back to Buenos Aires and you fold into Tanaka and I write a master’s report about a moon I never touched. And if I do, and we get caught, it’s your integrity file and my access revocation and both of us are done, and the only difference is which one of us gets blamed for it.” She picked up her chopsticks again. “I want it in writing that I understood the difference.”
Sanae looked at her for a long moment. Outside, Sagamihara was doing what Sagamihara did at eleven at night, which was nothing, under streetlights.
“I’ll put it in writing,” Sanae said.
She did not put it in writing. She meant to, on Friday, during the packet, and the packet took fourteen hours and ended with Miriam running the four bulk replicates on the cold line at two in the morning while Sanae formatted figures, and by Saturday there was no time and no good hour and the form stayed where it was, which was in a folder, unfiled.
Saturday was the seventeenth of October. Rin’s code opened the door at 2:14 a.m. The access log noted the code, the timestamp, the instrument: NanoSIMS-2, scheduled block, four hours. The log does not have a field for who is standing next to the person whose code opened the door.
The grain was S1-0417. Twenty-two microns, blocky olivine, catalog mass 8.7 nanograms estimated from the imaging stack, Site 1, Stickney ejecta unit, Mars-facing hemisphere. Rin loaded it with the tweezers under the microscope while Sanae watched the reticle and said left, more left, there in a voice she did not recognize as her own. The primary beam went on at 2:51. The crater opened at the rate of a micron an hour, or the speed of a glacier, or the speed of a career, depending on which unit you measured in.
The number came off the instrument at 5:40 a.m., after the standards, after the normalization, after Rin’s hands stopped shaking enough to operate the peak-jump routine.
Δ17O = +0.34 ‰, ±0.09.
Above the terrestrial line. Up with Mars. Three tenths of a per mil, plus a tenth of a per mil of honest uncertainty, floating in the empty upper field of a plot that had a shaded box in it labeled MARTIS MANTLE and a tick mark at +0.32 where the shergottites lived.
Sanae stared at it until the numbers went soft, and then she stared at it some more, and then Rin said, “Is that good,” and Sanae said, “It’s honest,” which was the most she could say, because it was one grain, and one grain of Mars-like oxygen in Mars-facing regolith was exactly what both hypotheses had always predicted and therefore settled nothing, exactly nothing, not a single degree of the argument, and she knew it while she looked, and she looked anyway, like a woman re-reading a pregnancy test she has already decided not to trust.
They left the grain in the sample holder, cratered, spent, a lunar-grade paperweight now. Rin took the train home at six in the morning with her student ID in her pocket and her code in the log and her name nowhere at all.
On Monday morning, Nao Miyake found it, as she found everything, by reconciling a count.
The weekly particle audit takes twenty minutes. Trays come out, trays go in, the handheld reader talks to the catalog, and the numbers fall down a column and either they agree or they don’t. On this Monday, tray S1, bin 20/25 microns, did not agree. Expected 418 catalogued particles. Found 417. The missing number was S1-0417, twenty-two microns, blocky olivine, Site 1, Stickney ejecta unit, Mars-facing hemisphere.
The catalog entry for S1-0417 had been updated by no one. The bench log for the isotope lab had no entry. The NanoSIMS-2 session file, however, was sitting on the server with a timestamp of Saturday, 2:51 a.m. to 6:02 a.m., and an operator code, and the operator code was KUBO.R.
Nao sat with this for a while in her office with the door shut. Outside, the facility was doing its Monday: someone in the hallway talking about a courier, the air handling cycling, the particular low thrum the nitrogen line made when it topped off a cabinet. She pulled the session file. She looked at the calibration standards, which were good. She looked at the sample identifier, S1-0417, and the count reconciliation, 418 to 417, and the operator code, KUBO.R, which was a twenty-four-year-old who had been in the building for fourteen months and who had, until this weekend, never once filled out an access request without a supervisor’s initials.
Nao pulled the access log next. Two-fourteen in the morning. Four-hour block. Student certification, active, no supervisory override on file, because the system did not ask for a supervisory override and if it had, the answer would have been typed by the same person typing everything else.
She knew. She knew it the same way she knew the nitrogen percentage by smell, without instruments. She opened the incident report form on her terminal, the one the joint protocol required her to file for any unauthorized expenditure of catalogued material, and she put the date in the first box, and the date was October 21, and she looked at the date for a while, and then she closed the form without filling in the second box, and she opened the bottom drawer of her desk, and she put the form in the drawer, under the intake agreement, page four, paragraph two, the one about the American entitlement, which she had read so many times she could recite it and which she kept where she kept it because it was the only document in the building she had never once had to sign.
Here is what the incident report would have done, if she had filed it on Monday morning: it would have gone to the joint review board, both partners, all at once, by protocol, without her discretion in the loop. It would have named Rin Kubo. It would have named Sanae Ozaki by inference, because the only living person with a motive to spend one grain of Site 1 on a Saturday morning in review week was Sanae Ozaki, and every board on two continents knew it. It would have locked the cabinet pending audit, which meant no sessions, no requests, no November agenda, nothing, until a committee of eight people who had never held a grain finished reading transcripts. It would have ended the October review panel before it opened, because an institute does not keep a group that is under integrity referral, and it would have ended Nao’s nineteen years of clean custody, because a collection with an incident in its file is a collection with a fingerprint on it.
And there was the number, sitting on the server, unfiled.
+0.34.
Nao was not a geochemist. She had curated geochemists. She knew what +0.34 meant, the same way a librarian knows what a rare book is worth without ever having read it: it was up with Mars, it was one grain, it was not a population, it was implantation or provenance or the first crack in a wall, and it was the kind of number that a woman with four people and a October twenty-second could not be blamed for spending a Saturday to get.
She closed the session file. She left the access log exactly where it was, unannotated, unreferred, in its normal place in the weekly folder, where anyone with curator credentials could find it and no one without reason would look. She did not call Rin Kubo. She did not call Sanae Ozaki. She put the weekly reconciliation in the drawer with the form, 418 to 417, initialed, dated, in her own handwriting, which was the only handwriting on that particular piece of paper, as it was on every other piece of paper in the building.
Then she went downstairs and stood in front of the cabinet for a while, and then she went home, and slept badly, and was back in the building at four Wednesday morning because she had never in her life slept well the week before a panel, and because the fluorination line’s roughing pump had been making a sound she did not like, and because she was, in the end, a woman whose job was to go and check things at four in the morning when they made a sound she did not like.
The isotope lab was lit, as labs are at four in the morning, which is to say entirely, because nobody had turned anything off. Sanae was at the NanoSIMS workstation in the corner with her coat still on, and the plot was on the big monitor, and the big monitor was the brightest object in the building.
Nao came in and stood behind her. Neither of them said anything for a while. On the screen, the terrestrial fractionation line ran diagonally across a field of gray, and the Mars field sat above it, a shaded band, and the carbonaceous field sat below, another shaded band, and in the upper band, alone, with crosshairs on it like a sighted rifle, was a single point with an error bar.
+0.34 ‰, ±0.09.
“One grain,” Nao said.
“One grain.”
“Site 1. Mars-facing.”
“Where both hypotheses have always said you’d find one.”
Nao looked at the screen for a while. The error bar was honest. The standards on the side panel were clean. The session file had a name on it that was not the name of the person sitting in the chair.
“Seventy-two hours to your panel,” Nao said.
“Sixty-one.”
“What are you going to tell them.”
Sanae spun the chair around, and there she was, at four in the morning, with her coat still on and her eyes the color of the room, and she looked at the curator of the collection for a long moment and did not say the thing that both of them could hear, which was: I could tell them you found the log on Monday and chose not to file it, and then neither of us would have to have this conversation, and my group would be finished by Thursday lunch.
“What you just said,” Sanae said, “about the board and the audit and the one percent on page four. I’ve been thinking about it all week. That if Halvorsen wanted tomorrow, he could have sixty-one milligrams by right, and the only thing standing between him and the order is that he hasn’t asked, because asking would mean he doesn’t already know.”
“Yes.”
“That’s the whole game, isn’t it. Curation. You keep the grain so the person who hasn’t been born can ask the question the person born today is afraid to write on a form.”
“I wouldn’t put it that way.”
“How would you put it.”
Nao thought about it. On the screen, the single point sat in the gray, above the line, with its error bar open like a hand.
“I keep it so that somebody in thirty years can call somebody in this room a fool,” she said, “with better instruments. That’s the job. Everything else is a signature.”
Sanae turned back to the screen. The plot glowed. Neither of them moved to turn it off, and neither of them mentioned the access log, and at 4:40 Nao poured two cups of the terrible coffee from the break room machine and put one on the corner of the workstation without comment, and went back to the roughing pump, which turned out to have been making exactly the sound it always made, which she knew already, and had known before she came downstairs.
The panel sat on Thursday the twenty-second, in a conference room in Sagamihara with a view of nothing in particular, and Sanae brought them the eleven milligrams and one grain, and nothing else, and she put them on the same slide.
Miriam had run the four bulk replicates on Tuesday and Wednesday. They came back at Δ17O = -0.28 ‰, ±0.01, on the Site 1 fines, four independent charges, ten thousand grains each, all of them below the terrestrial line, all of them down with the carbonaceous material, all of them clean and cold and exactly where the Tucson preprint said the body should sit. And there, beside it, in the lower corner of the same slide, the single grain: +0.34 ‰, ±0.09, one number, one crater, one morning.
“You’ll note,” Sanae said, “that these two numbers are not the same number.”
Chair Sundqvist, who had sat on enough of these panels to know when someone was about to spend her last ten minutes carefully, said: “Explain the discrepancy.”
“There is no discrepancy. There’s a mean and there’s a grain. The mean is four charges of homogenized fines from the most gardened ground on the body, averaged over ten thousand particles before the oxygen reached the mass spectrometer, and it lands with the carbonaceous field, which is what Dr. Halvorsen’s group predicted from orbit and published in August and has been cited forty-seven times. The grain is a single twenty-two-micron olivine fragment from the Mars-facing hemisphere of a tidally locked moon, and it lands with Mars, which is what implantation predicts, which is what delivery flux predicts, which is also what the giant-impact hypothesis predicts, and I am here to tell this panel that a mean and a grain are two different instruments pointed at the same question and neither of them has answered it.”
“And the origin question,” Sundqvist said.
“Is not answered. The bulk mean is consistent with capture. It is also consistent with a mixed regolith whose Mars-flux component has been diluted below the resolution of bulk averaging, which is a sentence I can support with arithmetic if the panel wants it, and which Dr. Halvorsen’s preprint does not discuss because his sample was an orbit.”
“And what,” said a man on the far side of the table whom Sanae did not know and did not like the look of, “would resolve it.”
“Population statistics on individual grains. Fifty catalogued grains per site, NanoSIMS, both sites, with the Site 2 reserve end-member released under witness signatures. That’s my November request. It’s on the board’s agenda with Dr. Miyake’s name on the sponsorship line.” She paused. “I want to be clear with this panel about what I did and did not do to get the number you’re looking at. The single grain was measured on Saturday the seventeenth under a valid student certification, in a properly calibrated session, with clean standards. The authorization chain for that session is incomplete. I’m not going to stand here and tell you it was complete.”
The room got quiet.
“The incomplete authorization is in the curator’s hands,” Sanae said. “Dr. Miyake has not chosen to act on it, and I’m not going to speculate about her reasons from this chair. What I will say is that the grain is real, the standards are clean, and the number is the number, and if this panel folds my group on the twenty-second, that number goes to Tanaka’s lab with the rest of us and Halvorsen keeps his citations until somebody with fifty grains and a November agenda takes them away.”
Sundqvist looked at her for a while. Then the panel went into the part of the meeting where they talked among themselves, and Sanae sat with her hands folded on her notebook, and Miriam’s knee under the table was not shaking at all, and Rin, at the far end, was taking minutes in a handwriting that would have embarrassed her.
The decision, when it came, took ninety seconds. The fold was deferred. The group’s status was extended to the November board, contingent on a two-site population dataset on the table by the ninth. If the dataset did not exist on the ninth, the fold was automatic and no further hearing was required. Nobody said the word probation. Nobody had to.
In the hallway afterward, Rin said, “You told them about the authorization.”
“I told them it was incomplete.”
“You didn’t say whose.”
“It wasn’t yours,” Sanae said. “It was mine. Yours was a valid certification used in a valid session by a supervisor who should have countersigned the form and didn’t. The incomplete authorization is the supervisor’s. That’s the file. That’s what I told the panel.”
Rin looked at her for a while in the hallway, under the fluorescent lights, with her minutes clutched against her coat, and then she said, “Thank you,” in a voice that did not thank her for the file.
On Friday morning, the twenty-third of October, Nao Miyake came into the isotope lab at 8:40 with a tray in both hands, wearing gloves, moving as she always moved in front of material, slowly and without wasted motion, the movement of a person who has spent thirty years being watched by her own signature.
She set the tray on the pass-through. The label read: SITE 2, TRAY D, RESERVE END-MEMBER, 30/40 MICRONS, CATALOGUED 2,216, INTAKE SEALED, DO NOT OPEN WITHOUT CURATOR AND PI COUNTERSIGNATURE. Site 2. The anti-Mars hemisphere. The undisturbed lag deposit from the fresh crater on the trailing side, the ground that faced away from Mars, the material that had spent four billion years not standing in the rain.
Sanae came out of the instrument room with her coat off and her sleeves up. She had not slept. She looked at the tray and did not touch it.
“I’ve been thinking about your Monday reconciliation,” Nao said.
“My what.”
“Tray S1, bin 20/25. Four-eighteen to four-seventeen. Initialed. Dated. In the weekly folder with everything else.” Nao opened the cabinet’s inner seal with the curator key and folded it back. “The number on this label is two thousand two hundred and sixteen. When I take a grain off it, it will say two thousand two hundred and fifteen, and I will initial it, and date it, and put it back in the cabinet, and that will be the file.”
“The file.”
“The file,” Nao said. “You want fifty grains per site, and the board will want to know why the reserve came open, and the answer I’m going to give the board is that I spent one grain on a screen because the screen was worth more than the grain, and I need the grain’s number before I can say that in a room with Halvorsen’s institution in it.”
She reached into the tray with the tweezers. The tweezers were curved-tip, antistatic, the same pair that had loaded every grain in the building since before Sanae’s group existed. The nitrogen hissed around her wrist. She worked down through the catalog order as she always did, patient, unhurried, reading numbers, and then the tips closed on a grain at the middle of the sheet and she lifted it out into the light.
It was thirty microns across. Gray, angular, dull, indistinguishable from three thousand of its neighbors and from every other grain Sanae had ever held under a microscope, and Sanae looked at it for a moment and understood that she was never going to be able to explain to anyone, in any room, on any panel, what she was looking at, and that this was in fact the only thing in nineteen years of her professional life that she had ever been unable to explain.
“If it comes back Martian,” Nao said, holding the grain over the sample holder, “the anti-Mars hemisphere is Martian, and implantation is off the table for the reserve end-member, and the release goes to the November board with my signature and a witness and no argument from Halvorsen’s institution that can survive contact with the arithmetic.”
“And if it comes back carbonaceous,” Sanae said, “then the reserve end-member is carbonaceous, and the Mars-like grain in Site 1 is a delivery-flux grain, and your bulk mean stands, and my group folds on the ninth and Tanaka gets four people and a very interesting master’s report.”
“Those are the two cases.”
“Those are the two cases.”
Nao set the grain in the holder under the microscope stage and did not look up. On the pass-through between them lay the reserve expenditure sheet, the one that had been in a folder in a drawer in her office since intake, blank except for the header, waiting for two names: CURATOR, AUTHORIZING, and PRINCIPAL INVESTIGATOR, RECEIVING. The first line already had Nao’s name on it, because Nao’s name was on everything, because that was what she was doing in the room, that was the whole of her immortality, a woman’s handwriting on the only piece of paper anybody would still be able to read in a hundred years, and she knew it, and she had known it since Monday, and she had come downstairs at four in the morning for the pump, and there had been nothing wrong with the pump.
She slid the sheet across the pass-through.
Sanae picked up the pen. The fluorination line behind them was cold, its brass manifold quiet, its roughing pump making exactly the sound it always made. The NanoSIMS would need an hour to warm and the standards would need another forty minutes after that and the session would not begin until nearly noon, and none of that was the point, and both of them knew none of it was the point, and neither of them said so.
Sanae signed the second line.
Nao took the sheet, checked the ink, folded it once, and put it in her coat pocket. Then she picked up the pen again, took the paper slip off the lid of Tray D, and wrote the new count in the blank at the bottom of the label, in the handwriting everybody in the building could recognize with their eyes shut: 2,215. Dated October 23. Initials. She taped the slip back onto the lid, pressed the corners down with her thumb, and carried the tray back to the cabinet, where the nitrogen took it without comment, as nitrogen takes everything, and the door sealed, and the count stood.