I'd been thinking in small, fast loops all morning — variables, safety margins, what could kill me and what could be turned into leverage. The chunk of metal I'd found the day before sat heavy in my palm again: gunmetal-gray, glassy fracture, a texture unlike ordinary iron. I tapped it on a flat stone out of habit.
Sparks.
Not a shower of fire like a grinder makes, but a stubborn, bright shower of orange that ran like a vein across the contact point. That meant something: carbon content, or at least something that would throw a spark when struck. Workable. Not worthless.
I tested a few other samples the same way. One piece—dull, red-brown—gave dull, sparse sparks. Another blasted a cleaner, whiter flash. My brain was a measuring tape and a scale without numbers: color of spark, how it shattered, how dense it felt in my fist. The gunmetal one felt dense for its size, like it used less space to be heavy. That, too, mattered. Whatever this world called ore, there were rules I could read if I jabbed at them hard enough.
I didn't have the luxury of a lab, so everything was low-tech hypothesis testing. Tap or rub on stone to check sparks. Scrape with a blade to see the true color beneath the patina. Bite the thin flakes (no, I didn't actually put metal in my mouth — spare me the lecture); instead, I rubbed metal on rock and judged the residue. Rust or verdigris came off, revealing the skin underneath: coppery? Steel-gray? Every little cue was a data point.
Outside, the river had been a smear of silver. The herd was still there when I checked the crack in the wall. I waited. Thirty minutes passed like an eternity of listening. Time is weirder when everything is a risk calculation. Ten more minutes and their bulk rolled away downstream—the current had called them, or something else. Either way, the river cleared.
Good. Not safe forever, but clear enough for a run.
First problem: clothes. Second problem: water. I could have stripped the grime off myself and chugged the river like an idiot, but I was not an idiot exactly. The memory of sterile labs had left me with a superstition in the form of knowledge: unseen microbes could end a man faster than a herd ever could. No way I was drinking whatever pooled by animals without thinking.
I carved a bowl.
I found a slab of deadwood—heartwood, dense and dry, not the sapwood that would rot in a week. Using the dagger, I hacked shallow cuts and worked a hollow with scraping motions. The blade complained; my palms burned. The hollow didn't have to be pretty. It had to hold water. I scraped until the cavity had a lip and could be carried without sloshing. Then I roasted it lightly over coals to tighten the fibers and drive out sap that might leak. That small roast smelled like burnt sugar and bark and felt goddamn satisfying.
Next: water. I scooped a shallow amount and thought through sterilization like a checklist. Boiling is the standard. I had no pot worth trusting, only that crusty metal pot crusted in soot and a cracked crucible. But you don't need a fancy kettle to kill pathogens; you need heat. Heat can be brought to water by hot rocks. Old technique. It's clumsy, slow, but it works.
Picking the right rocks is a science if you care about not fracturing the damn thing in your hand. Porous or fissured stones explode when they meet heat and moisture. I picked river stones—smooth, dense, dark. Tap them; they rang dull and solid. No hairline fractures. Good.
I gathered tinder: dried moss from a sheltered root, resin oozing on a pine limb (sticky and gold), brittle twigs, leaves that crumpled to dust under my fingers. The resin was the real starter — drizzle a little and it will catch easier than any blown tinder. I arranged kindling in a teepee over a small depression in the ground, laid the resin like a fuel line, and struck sparks with two flint-like stones until the resin let out a sizzle and then a thin blue tongue of flame.
Fire is etiquette with physics. Feed it small, then bigger; keep air moving, don't smother. The bellows here was toast, but that didn't stop me. I cupped my hands, blew a measured breath, and coaxed the flame into stubborn life. Sparks catch, resin blazes, sticks smolder, then surrender into a heat I could use.
I fed in the smallest river stones first to heat, then the bigger ones. When the rocks glowed dull red I plucked them with two split branches, careful to balance the weight and avoid a blister. Into the wooden bowl they went. Steam hissed like teeth. Bubbles swelled at the edges. I repeated the cycle three times—hot rock, in the water, let it sit until it steamed, pull it out, heat another bunch—until the water surprised me into a steady, rolling heat that told me pathogens were, for the most part, done. No thermometer, just a feeling and the science behind it: sustained near-boil equals sterilized.
I kept some water aside, wrapped in a cloth and tucked into a shaded hollow. Conserving is an art I'd learned in theory and now practiced. A litre or two could be the difference between clear thinking and stupid desperation. The bowl sat at my feet like a small victory.
Back at the smithy, I turned back to the instruments. The crucible was crusted; the inside held a film of old slag and rust. I could throw it away, but scrap is valuable when you plan to build. Clean first, test later.
I made a quick sand-lather: sand grabbed from a riverbank, slightly damp, and a stick to rub. It did the job, abrasive to grit the old crust loose. But sand and elbow grease are slow, and they take surfaces off unevenly. I could see the difference between what I could clean by brute force and what I could make by invention.
That's when an idea landed. It felt stupid initially, then immediately like the kind of stupid you double down on.
Sandpaper. The simplest abrasive tool humanity had probably made since the first caveman filed a bone.
Make a flat backing. Glue abrasive particles to it. Done.
Problem: backing and glue. There were skins in the world—leather was ideal, but no carcasses within reach. The inner bark of some trees peels into thin, flexible sheets. I found a strip: damp, pithy, but resilient once dried. I scraped it with a blade until it lay thin and pliable, then dried it over coals until it became a papery sheet.
Glue was easier than I expected. Resin when heated liquefies into a syrup that can stick sand like glue—no chemistry degree required. I heated a clump of resin until it softened, added a pinch of charcoal ash to thicken and give some tooth, and stirred with a twig until it became a tacky paste.
I spread a thin layer across the dried bark with a flat stick, sprinkled river sand in even strokes, and pressed the sand in with a flat stone until it seated. The surface looked ugly and primitive, but the feel under my thumb was rough in the best way: aggressive grit, even coverage.
First test: the dagger. I took the edge and rubbed it along the new abrasive in controlled passes, feeling the hairs on my neck stand up as the metal cleaned, then brightened. The rust came off in ribbons. The edge grew keener. Balance stayed even. Handle comfort was marginal—old wood, not ergonomically carved—so I sanded the grip smooth, rolled it in resin, and wrapped a strip of bark for friction.
Then the crucible. The sandpaper moved through the slag like a small miracle: the abrasive scoured the interior rim, loosened the flakes, revealed dull gray metal under the old crust. I used sand and my new paper to polish the interior until I could see the true lip of the vessel.
It felt childish and monumental at once. Something so small — bark, resin, sand — had turned into a tool that would change everything. It was not some romantic, world-bending device; it was deliberate, boring, reliable. That's the kind of invention I liked. Practicality kissed elegance.
By dusk the daggers were sharper, the crucible cleaned, a small cache of stones lay beside the hearth ready for the next boil, and my clothes hung on a drying branch, smelling faintly of resin-smoke but cleaner than before. I had water that would not kill me and a sandpaper that would make metal obey.
I sat and let the night fill the world with a slow, steady hum. In the dark, the sounds of the forest felt closer and less alien. The campfire pushed out a circle of safety. Tomorrow I'd see how the cleaned crucible handled a real melt—whether the gunmetal chunk could be coaxed into bloom or whether the copper powder would behave like powder and not slag. Tomorrow I'd test my little tools against the rules of this world.
For now, I packed sandpaper and tools into a corner, wrote the steps of the process in my head like a recipe, and felt the tiniest, stupidest grin curl my lips. First invention: sandpaper. First victory: practical and honest.
The workshop smelled like burnt bark and river grit. It smelled like beginnings.