Monday arrived the way Mondays always arrived: with the grey preliminary light of February coming through the north window at six in the morning, Halvern Ridge still blocking the direct sun and the workshop sitting in a darkness that was not quite night and not yet day. The kind of light that required a lamp. The kind that made things visible without making them bright.

Corin opened the Seven-Day Ledger.

He sat at the small table by the window -- not the workbench, not yet, he never went directly to the workbench until he had established some sense of what the week would be -- with the Ledger open and the pen in his right hand and the entries above the current page still reading the same as they had when he had written them seven days ago. Cycle number. Date. The approximate time of transition. He read them over once, as he always did, establishing the continuity: there had been a person sitting at this table seven days ago, and the person sitting here now was the same person in every way that could be confirmed by record.

He wrote: Cycle 74. Started Monday, 02:14. Transitioned from: bookbinding (Cycle 73). Preliminary orientation: unknown, pending assessment.

Then he set the pen down and went to find out what the assessment would say.

---

The geological samples were in the lower drawer of the workbench, where he had placed them two weeks ago when mineral analysis had last been the week's orientation and he had been partway through an assessment of the Halvern survey cores. Three sections, each sealed in a small archival bag with his handwriting on the label: date, site location, depth. He took out the largest, the one with the highest concentration of Grey Ferrite inclusions, and held it in his left palm.

The information arrived through his skin before he consciously registered receiving it. Temperature differentials across the sample's surface. The weight distribution, which was not uniform -- the Grey Ferrite seams were denser than the surrounding matrix stone, creating a subtle asymmetry that his hands had last registered two weeks ago and now registered again as if no time had passed. The crystalline structure, telling him things about formation conditions and geological age that he had spent six days learning to read two cycles back and which his hands remembered even now, in the attenuated way motor memory survived in the intervals between its full manifestations.

He set the sample down.

Mineral analysis. Again. He had suspected it from the way the geological samples had felt yesterday evening -- a gathering quality in his hands as they handled the labeled bags, a preference that was not yet competence but was the precursor to competence. His nervous system had been listening to the rocks.

He returned to the Ledger and wrote: Current cycle: mineral analysis, geological focus. Incoming competence: Day 1, standard rate. Physical symptoms: shoulder stiffness (expected, subsiding), right forearm sensitivity at transition site (mild). No complications.

He read what he had written. Then he looked at the bookbinding tools on the shelf above the workbench: the bone folder in its usual place, the needles rolled in their case, the backing hammer on its hook. He had used these tools every day for a week with a fluency that his hands could not now reproduce. He went to the shelf and took down the bone folder.

He held it in the position his memory accurately described. He made the motion -- a simple paper fold, the most basic application of the tool, the kind of thing that should require no thought. The fold was wrong before he finished making it. Uneven pressure, a slight buckle at the midpoint, the result that a person who had never been taught bookbinding might produce rather than the result that six days of complete motor competence should produce.

He completed the fold. Looked at it. Set the bone folder on the shelf.

This was the Monday ritual he had never explained to anyone: testing the previous week's skill to confirm its absence, not because the test changed anything but because ambiguity was worse than the loss itself. He needed to know clearly what the week was and what it was not. He had learned this from experience: the first several Monday mornings of his condition, he had spent hours trying to recover what had left, working against the absence until he was exhausted and frustrated and no closer to anything. Now he confirmed the transition cleanly and moved on. The bookbinding was gone. The mineral analysis was present. He was Cycle 74.

He put the bone folder back in its correct place.

---

He made coffee -- too dark, the consistency of tar, the way he had always made it and which was probably one of the things Tessa considered a character defect -- and stood at the north window while it cooled enough to drink.

Outside, Halvern Ridge filled the upper half of the view. At this hour in February the ridge was grey-black against a grey sky, the snow line somewhere above the tree line, the peaks lost in cloud. It looked exactly as it had looked at six this morning seventy-two cycles ago, which was when he had first developed the habit of standing at this window after the Monday assessment: looking at the mountain the way you looked at the source of a thing you were trying to understand.

The Grey Ferrite deposits ran through the ridge in seams that the old mine maps showed as reddish threads through the grey stone. The same mineral that had been in the walls around him for six hours in the dark of the collapsed tunnel, eighteen months and seventy-four cycles ago. Dr. Rosch's working hypothesis -- stated carefully, repeatedly hedged, never presented as more than provisional -- was that the combination of the Grey Ferrite's electromagnetic properties, the crystalline dust that had settled on his skin and respiratory system, and the prolonged hypoxia during the wait for rescue had created an unusual and as-yet-poorly-understood interaction with the neuroplasticity mechanisms of his nervous system. The result was what it was: every 168 hours, plus or minus a few, his motor memory reassigned itself around whatever domain his attention had been most concentrated in the preceding days.

Hyper-accelerated procedural learning, Rosch called it. With the particular cost of non-retention.

Corin had stopped spending very much time thinking about the mechanism. The mechanism would reveal itself or it would not, and his thinking about it had not proved particularly useful. What he spent his time on was managing the practical reality: the work that was possible during each cycle, the transitions, the records, the keeping track of what he knew and what he had known and what would eventually return.

He thought about it sometimes as a kind of professional occupational hazard. He was a conservator. His entire working life was oriented around the proposition that what had been made and used and damaged and partially lost could be preserved, could be brought back to functional stability, could be given the chance to survive into the future rather than disappearing into the past. He was professionally committed to the idea that things worth keeping should be kept.

And every Monday morning he stood at this window with different hands than he had stood here with the previous Monday, having lost what the previous week had most fully made him.

A conservator who could not conserve himself.

He had written that in the margin of the Ledger once and then crossed it out because it was the kind of formulation that was accurate enough to feel true and too neat to actually be useful. The truth of his situation was not neat. It was not an irony that could be stated and then set aside. It was the daily texture of his life, which he moved through the same way he moved through everything else in his life: precisely, without sentimentality, doing the work that was in front of him.

He drank his coffee. The geological sample was on the workbench, waiting.

---

At Archive House, mid-morning, he spread the samples across the cataloguing room table alongside the reference materials and the preliminary notes from two weeks ago, when this cycle had last been present. The work picked up exactly where it had left off in the way that all his returning cycles did -- not seamlessly, because the re-acquisition period on Day 1 was always slightly rough, but with the quality of recognition rather than relearning. His hands had done this before. His hands knew this.

The trace element pattern was the same as it had been two weeks ago, which was what he had expected. What he had not expected was that the cross-referencing he had done with the standard geological reference texts had left a gap he had not noticed before -- a specific combination of silica and iron and the third element that kept appearing in these samples without a matching profile in any of the regional sources. He had been going to look at that more carefully, and then the bookbinding cycle had arrived, and the mineral attention had stopped.

He was looking at it now.

The combination did not match any ore source in the standard reference literature. That had two possible explanations: the ore source was genuinely unusual and had not been previously documented, or the reference literature was incomplete in the region and this combination was standard for Halvern-area ore. He needed more samples to determine which.

He was writing up the updated preliminary assessment when he heard the door of the building open on the ground floor, and then footsteps going up the stairs to the reading room. He did not look up. Tessa had mentioned a researcher arriving.

He worked through the morning with the specific focused quality that mineral analysis weeks brought -- patient with inconclusive data in a way he was not patient during other cycles, willing to sit with a question that did not yet have an answer and trust that the answer would become visible if he kept looking carefully. This was the epistemological style of geology: you looked at the rock, and the rock told you what it had been through, and you trusted the rock's record over your own theories about what it should say.

Tessa appeared in the doorway of the cataloguing room at half past eleven.

She stood there for a moment before speaking -- a habit he had stopped noticing in the same way he had stopped noticing the river. "She's upstairs," Tessa said.

"I know."

"She had a great many questions about the smelting records."

He looked up from the preliminary assessment. "What kind of questions?"

"The kind that suggest she already knows what she's looking for and wants to verify it against the primary sources." Tessa considered something briefly. "She's thorough. The kind of thorough that is methodical rather than comprehensive. She's not trying to know everything. She's trying to confirm a specific hypothesis."

Corin thought about this. Researchers who already had a hypothesis when they arrived were either very good or very fixed in their thinking. The difference between those two things was whether the hypothesis changed when the evidence suggested it should.

"What's the hypothesis?" he asked.

"She hasn't said it explicitly. But the questions are about cross-guild membership." A pause. "The frequency of it in the records."

He looked at the geological samples on the table. He looked at the samples for long enough that Tessa's silence became a question.

"I'll be up tomorrow," he said.

Tessa received this information with her customary economy. "She'll be here most of the week," she said, and went back to her desk.

Corin returned to the trace element pattern. Outside the high window of the cataloguing room, Halvern Ridge was beginning to disappear into the low cloud that came down most afternoons in February, the grey peaks going grey-white, the boundary between rock and sky dissolving. A researcher upstairs asking questions about cross-guild membership. He thought about what she might find in those records, and what she might make of it, and whether any of it could come anywhere close to what he knew from the inside.

He wrote in his notes: Ore source genuinely anomalous OR Halvern-specific profile not in standard literature. Next step: compare against Halvern mine survey geological reports, pre-accident files.

He underlined pre-accident files.

Three months, Tessa had said. He supposed he would find out.

He stayed in the cataloguing room until the archive closed at five, working carefully, and did not go upstairs once.

There was one more thing about Monday mornings that he had never told anyone, not even Dr. Rosch, because it did not fit neatly into a medical framework and he was not sure what framework it did fit into. On Monday mornings, in the brief interval between the transition ending and the new cycle fully establishing itself -- the transitional hour, the hour when neither week's skill was fully present -- he sometimes felt something that he did not have a name for. A quality of openness. Not a pleasant feeling, exactly, not comfortable, more like the feeling of a door left open in a room that was supposed to be closed. As if the self that organized around skill and competence and the useful deployment of capability had briefly, in the interval, no particular shape at all.

He was not that person in those moments. He was not the bookbinder, not yet the geologist, not the conservator who had a professional reputation and a workshop and a list of things that were still his. He was simply someone sitting at a table in the grey early morning, holding a pen, looking at the mountain.

He had spent considerable time, in the early months, trying to determine whether this feeling was frightening or important. He had concluded, over time, that it was neither -- or rather, that it was both, and that those two things were not opposites. The interval between one version of himself and the next was the most honest version of himself there was, and that was not frightening in the way that danger was frightening. It was frightening in the way that the truth was sometimes frightening when you had been successfully avoiding it.

He did not know what that meant about who he was outside of what he could do.

He thought Rosch might have a framework for that. He had not asked.

---

By the time he left Archive House that evening, the preliminary assessment had yielded two things worth noting. The first was the trace element profile, which was anomalous in a way that was beginning to look less like incomplete reference literature and more like a genuine geological singularity -- something specific to the Halvern ore that had not been documented because the people who had worked with it had not been geologists, had known the ore by its properties rather than its composition. He needed the pre-accident survey files to confirm this. They were somewhere in the archive's collection, or in the mine company's records, which Tessa had mentioned were in the archive's custody since the company had dissolved.

The second thing was harder to write up in the preliminary notes, because it was more in the category of observation than conclusion. In the course of the afternoon he had become aware, at intervals, of footsteps in the reading room above. Of a voice asking Tessa something in the hallway outside the cataloguing room -- not the words, just the tone: precise, unhurried, the tone of someone who had prepared their questions carefully and was asking them in order. Once, there had been a pause in the movement above him that had lasted long enough that he had, without meaning to, stopped writing and listened.

Then the movement had resumed.

He did not know what the researcher upstairs was looking for. He knew what was in the Smiths' Guild records for the period Tessa had said she was examining. He knew what it looked like when you found the cross-guild entries for the first time. He had found them himself, eighteen months ago, in a different cycle and a different context, in the early months when he had been trying to understand what had happened to him and had started by trying to understand the place.

He had not found a complete answer. He had found the records, and the records had asked as many questions as they had answered, and he had set them aside because the questions they asked were not the kind he was equipped to answer alone.

The researcher upstairs was apparently equipped, or thought she was.

Three months, Tessa had said.

He locked the cataloguing room and said good night to Tessa and went out into the February evening, and the cold settled around him the way cold always settled in Alderheath -- thoroughly, without particular cruelty, just as a condition of the place and the season and the proximity of the ridge that made its own weather and sent it down into the valley at whatever temperature the mountain decided.

He did not look up at the reading room windows on his way past the front of the building.

He thought about the trace element profile. He thought about the pre-accident files. He thought about three months and what a researcher looking for something in a cross-guild pattern might eventually find if she kept looking.

He walked home through the grey evening and went to work on the geological samples, which knew what they were and were not keeping it secret.

The samples waited on the workbench in the lamplight, patient as stones always were -- which was very patient indeed, patient on a geological scale that made human patience look like impatience dressed up in a longer coat. He picked up the largest one and turned it over in his hands. Mineral analysis week. Whatever the ridge had to say, this was the week he had the best chance of hearing it.