ㅤ
By Thursday morning, Mira had decided Evelyn's next visit required a seating plan.
ㅤ
She drew four circles around our kitchen table, labeled three of them with crooked letters, and gave Sir Flopsy the chair nearest the window. She had moved the chairs twice already, although the table itself had not changed.
ㅤ
“That's where the light is,” she explained.
ㅤ
“Does he need light to eat imaginary supper?”
ㅤ
“He needs to see everyone.”
ㅤ
I moved the pan away from the edge of the stove. “I haven't invited Evelyn yet.”
ㅤ
“You said you could.”
ㅤ
“I said I could ask. Those are different things.”
ㅤ
Mira looked at her drawing, then added a question mark beside Evelyn's circle. She'd learned to make those at Mrs. Fenwick's and was using them whenever she thought an adult had left something unfinished.
ㅤ
“I'll ask her today,” I said.
ㅤ
She crossed out the question mark. Apparently that was sufficient. I put the seating plan somewhere the breakfast dishes could not reach it.
ㅤ
***
ㅤ
Halden's office had acquired two visitors by the time I arrived. One was Nessa, who had supervised the exhibition's safety cutoff. The other was a narrow-faced lecturer I recognized from the audience but had never met.
ㅤ
“The defensive studies seminar wants a demonstration next week,” Halden said, handing me a request. “A smaller room. No public audience. They'll be examining the method, not applauding the result.”
ㅤ
“That's reassuring.”
ㅤ
“It wasn't meant to be.”
ㅤ
The lecturer introduced himself as Professor Venn. He tapped the line listing the proposed attendees.
ㅤ
“One of the visiting researchers has asked whether the method can be adapted to protect a moving target.”
ㅤ
“We haven't tested that,” I said.
ㅤ
“Could it be?”
ㅤ
“Possibly. Movement changes where the strain gathers. We'd need to measure it.”
ㅤ
Venn turned to Halden. “Will there be a moving-target demonstration?”
ㅤ
“Not unless you want a demonstration of an untested configuration failing,” Halden replied.
ㅤ
Nessa cleared her throat. “If they do test it, I'll supervise the cutoff. But not as part of the seminar's advertised result.”
ㅤ
Venn looked as though he'd hoped for something more impressive. He left with a copy of the stationary test schedule. I checked that the request listed Tuesday of the following week, not an extra session that afternoon.
ㅤ
Halden gave me the request. “Discuss it with Evelyn. Both names remain on the presentation. And keep the seminar separate from whatever new trials you decide to run.”
ㅤ
***
ㅤ
I found Evelyn after her morning practical, standing beneath an archway with her notes balanced on one arm. She read the request while I explained the proposed audience.
ㅤ
“A moving target,” she said. “They want to know whether it can shield a carriage.”
ㅤ
“Or someone running from a carriage.”
ㅤ
“That too.”
ㅤ
Our exhibition had measured stability under controlled impacts. It hadn't established what would happen if the caster had to move, adjust their footing, and keep the barrier aligned with someone else.
ㅤ
“We could do a low-power trial today,” she said. “Not for the seminar. To see what we're missing.”
ㅤ
“Nessa has offered to supervise.”
ㅤ
“Then we can make a proper record of the failure, if there is one.”
ㅤ
“And keep the apparatus within the same limits as the exhibition,” I said. “I don't want to call a different experiment a comparison just because the diagram looks familiar.”
ㅤ
I waited while she signed the request beside my name. Her handwriting was precise even when she had nowhere to rest the page.
ㅤ
“Also,” I said, “Mira would like you to come for supper.”
ㅤ
Evelyn looked up.
ㅤ
“She made a seating plan. You're currently represented by a circle with a question mark that's been crossed out.”
ㅤ
“Am I expected to bring anything?”
ㅤ
“Only yourself. And perhaps patience. Sir Flopsy has been given a chair with a better view than mine.”
ㅤ
“Then I'll have to earn his approval.”
ㅤ
“Saturday? I can cook after my tutoring.”
ㅤ
“Saturday,” she said. “I'd like that.”
ㅤ
I had invited her because I wanted her at our table, not because Mira had prepared a diagram. Still, I was glad I could go home with an answer.
ㅤ
***
ㅤ
Thursday afternoon, Halden sent a message asking us to meet a visiting researcher in the practice room. She had requested a discussion of the exhibition report before deciding what to ask at Tuesday's seminar.
ㅤ
Evelyn arrived before me. A young woman stood beside her, studying the chalk diagram on the board. She wore a plain gray coat and carried a leather satchel with a broken buckle repaired in blue thread. Her brown hair was tucked beneath a modest hat. She looked much like any other visitor who'd walked a long way with a heavy bag.
ㅤ
“This is Miss Arden,” Evelyn said. “She's reviewing the seminar material.”
ㅤ
“Adrian,” I said.
ㅤ
“I know.” The visitor closed her notebook. “Your name is on the report.”
ㅤ
She offered no institution or family name beyond Arden. I noticed, but Halden had authorized the meeting, and I had no reason to demand an explanation at the door. I set our approved exhibition report on the desk and kept the loose pages from the new trial in my folder.
ㅤ
“I've read your report,” she said. “I'd like to understand why you kept the unsuccessful correction in the results.”
ㅤ
That was a better opening question than I'd expected.
ㅤ
***
ㅤ
Miss Arden began with the failed correction. The figure was small, tucked into the margin of our second calculation. It showed the point at which shifting mana between channels became more expensive than letting one section fail safely.
ㅤ
“Why publish that?” she asked. “It weakens the claim on the first page.”
ㅤ
“It narrows the claim,” I said. “Someone repeating the method needs to know when it stops helping.”
ㅤ
“Even if they follow your instructions?”
ㅤ
“Especially if they follow only the attractive parts.”
ㅤ
She made a note. Evelyn took the chalk and showed how we'd set the crystal's impact interval, then explained why the cutoff assistant had to watch the measurement dial rather than the appearance of the barrier.
ㅤ
Miss Arden asked how much of the result depended on Evelyn's control of the tests.
ㅤ
“Enough that you shouldn't separate our methods,” Evelyn said. “Adrian's model tells us where to redirect the load. My sequence tells us whether the redirection survives an impact.”
ㅤ
Miss Arden wrote that down too.
ㅤ
Then she turned to the figure on the board. “And if the protected person is moving?”
ㅤ
I drew a figure stepping sideways beneath the three-channel dome. The first problem wasn't speed. It was whether the shield moved with the person or held its shape around a fixed point. Either choice changed the mana cost.
ㅤ
“If I were protecting someone,” I said, “I'd rather use a shorter-lived barrier I could place correctly than a longer one that stayed where they used to be.”
ㅤ
“You make duration sound unimportant.”
ㅤ
“It matters. Just not more than the person.”
ㅤ
***
ㅤ
Nessa joined us for the trial we'd arranged. She checked the apparatus, marked the safe operating limit, and took her position by the cutoff.
ㅤ
The first pulse struck the barrier. I shifted the load. The second arrived while I was moving the shield's edge toward a chalk mark on the floor.
ㅤ
It collapsed early. Nessa had already moved her hand toward the cutoff; she stopped the next pulse before either of us attempted to restore the shield. I wrote down the point of failure while Evelyn checked the chalk target.
ㅤ
Miss Arden looked from the empty air to the diagram. “Was that expected?”
ㅤ
“I hoped for another second,” I admitted.
ㅤ
“Could you have held it with more mana?”
ㅤ
“Possibly. But we'd need a new baseline before comparing it with the exhibition result.”
ㅤ
Evelyn crouched to inspect the chalk mark. “The edge didn't reach the target. You moved the load too late.”
ㅤ
“I started when the first pulse ended.”
ㅤ
“Then we need to start before it ends.”
ㅤ
“That risks weakening the side still taking the impact.”
ㅤ
“I know.” She stood. “That's the problem we should test.”
ㅤ
Miss Arden watched us discuss the diagram. She didn't offer a solution. I was grateful; an answer from someone who hadn't handled the apparatus would only give us another assumption to check.
ㅤ
***
ㅤ
After Nessa shut down the equipment, Miss Arden asked whether she could keep a copy of the failed trial.
ㅤ
“The seminar isn't until Tuesday,” I said.
ㅤ
“I mean today's measurements.”
ㅤ
“They're preliminary,” Evelyn said. “We haven't checked the dial readings against Nessa's record.”
ㅤ
“Could you send them once you have?”
ㅤ
I hesitated. The report from the exhibition was available to the seminar attendees. Today's trial was not, and I didn't know who Miss Arden represented.
ㅤ
“I can ask Halden how he'd like us to share new data,” I said. “Evelyn and I need to check it against Nessa's record and agree on what it shows first.”
ㅤ
Miss Arden closed her notebook. “Of course. I should have asked who controlled the records.”
ㅤ
She didn't sound offended. That made it easier to ask my own question.
ㅤ
“Which institution are you with, Miss Arden?”
ㅤ
“A research office that advises on defensive applications. I'm attending the seminar as an observer.”
ㅤ
“Does the office have a name?”
ㅤ
“It does. I'm not here on its behalf today, and I don't want to imply that it has approved your work.”
ㅤ
A careful answer. It told me what she wasn't claiming, but not who had sent her.
ㅤ
“Then we'll see you Tuesday,” I said.
ㅤ
“I hope so. Thank you, both of you.”
ㅤ
***
ㅤ
When she left, Evelyn remained by the board.
ㅤ
“She spent more time on the failed correction than the successful result,” she said.
ㅤ
“That was the part she couldn't judge from the conclusion.”
ㅤ
“And she asked for today's data before we had checked it.”
ㅤ
“Then accepted no for an answer.”
ㅤ
“Yes.” Evelyn set down the chalk. “I don't think she was trying to steal it. I think she wanted to see whether we'd give it away.”
ㅤ
I thought of the way Miss Arden had withdrawn her request without argument. Evelyn's interpretation was possible, but we didn't know enough to settle it.
ㅤ
“She said she's an observer, not a representative. We should tell Halden exactly what she asked.”
ㅤ
“And what she didn't answer,” Evelyn said.
ㅤ
Neither of us had proof that her questions meant anything beyond professional curiosity. That was reason enough to report them accurately rather than supply a motive of our own.
ㅤ
She drew a line through my moving-target configuration.
ㅤ
“This part won't work.”
ㅤ
I looked closer. She was right. The revised load path would pull against the section still absorbing the first impact.
ㅤ
“Can you stay another ten minutes?” I asked.
ㅤ
“I can stay fifteen.”
ㅤ
We used all of them. We left the failed measurement labeled preliminary, with no revised claim for Tuesday's seminar.
ㅤ
***
ㅤ
At home, Mira asked whether Evelyn had accepted the invitation before I'd taken off my coat.
ㅤ
“Saturday,” I said.
ㅤ
She fetched the seating plan and replaced the crossed-out question mark with a flower.
ㅤ
“Does she like carrots?”
ㅤ
“I don't know.”
ㅤ
“You should ask her.”
ㅤ
“I will. We have time.”
ㅤ
I had not asked about carrots when I invited Evelyn. Mira had found the one gap in my preparations that mattered to her.
ㅤ
Mira set Sir Flopsy in his chair and went to wash her hands. I put the shopping list beside her drawing and added carrots with a question mark. When she returned, she pointed at it.
ㅤ
“Now you have one too.”
ㅤ
“Apparently I do.”
ㅤ
She took the pencil from me, crossed out the question mark, and wrote ask Evelyn beneath it in letters that leaned against one another.