Before I started experimenting, I visited the first floor to show my presence and keep the area secure, again with a cart filled with shells. This time, there were no surprises. The defensive walls were sprouting at an impressive speed that would make a Roman legion proud, while the others were already in place.

Everything was perfect. All I needed was to purchase more crossbows, and the corner of the first floor would belong to my budding guild. I could probably take over all of it, but expanding into the second floor was the better option.

If I took over the first floor and made a lot of money, it would either push the other guilds away completely or encourage them to explore the lower floors to make money like I did. Neither option was good for my objectives.

I returned to the fourth floor, my mind already churning on how to improve Nurture. The pattern of improvement was simple. I just needed to use the poisonous variant of the energy while I used my Nurture skill.

It was a simple plan, but there was one practical challenge. The only reason I was able to push the growth of the Nurture was the poisonous nature of the swamp water, which allowed me to push it further. However, just like any other challenge from the System's perspective, its improvement effect dropped off significantly after a certain point.

I tested to see if the tainted energy worked as a better poison. Unfortunately, it did not.

"Let's start doing some chemistry experiments," I said. When it came to chemistry, I knew even less than material science. The latter, I worked hard to educate myself after the Calamity, and working as a Blacksmith for three years gave me a lot of practical experience.

Chemistry, I knew nothing about. I couldn't even start guessing if the poison was some kind of organism or disease, a type of metal poisoning, or some kind of magic effect that bypassed the easy methods.

"Luckily, that doesn't mean I can't learn," I said. It was the whole point of science. Experiment and learn.

And, with the abundance of materials available, setting up the experimental setup was easy. I thought of three easy directions. Boiling, straining, and spinning. Three easy methods to get a general idea about the direction I should follow.

Boiling was straightforward. I just needed to get a container full of muddy water, boil it, and see if the evaporation concentrated the poison. Considering the number of open fires I had in place with the blast furnaces constantly running, preparing the necessary environment barely took a minute.

I even set up an angular metal plate over it to catch the steam and direct it to a secondary container, just in case the poison component moved with the steam.

While that boiled, I forged a number of metal sieves, each with different mesh sizes. For a few of them, I even went to the trouble of creating hexagonal gaps in case it had some kind of mana effect that would be affected.

Hexagon patterns didn't help, but the material choice did. Using silver somehow reduced the poison effect, while iron, even the anti-corrosive variant, didn't destroy it.

Third, I created a centrifuge. At least, a simple variant. The top-tier scientific centrifuges before the Calamity could rotate over a hundred thousand times per minute. I knew that, because I remembered one of the geneticists bragging for a whole meal about getting a grant to purchase one with an RPM of over two hundred thousand with some other specifications as well, with a price tag of over a million dollars.

Instead, even with my skills helping me, I was lucky to reach 300 rotations per minute on my current device. While my forging skills helped me to handle the material challenges, the real challenge came from the mechanical aspects. It was not as straightforward.

Still, it should be enough to work.

Once my setup was complete, I ran a set of experiments, which gave me three inputs. First, boiling destroyed both the fertilizer effect and the poison, leaving something closer to pure water. Not enough to drink it, but then the setup wasn't exactly designed for pure isolation.

A more rigorous distillation setup might have given me pure water. Not exactly critical under the circumstances, but nice to know in case of an emergency. It meant that even if I got lost, I wouldn't die from thirst.

This narrative has been purloined without the author's approval. Report any appearances on Amazon.

The second set of experiments was simpler. Using sieves was somewhat helpful. It wasn't exactly perfect, as the material on both sides contained both fertilizer and poison effect. The ground particles had worked more as a fertilizer, while the water acted as a poison.

I still put both materials on the centrifuge, which allowed the dirty water to coalesce even further. Once again, the water part carried the poison part. However, separating it from the water didn't exactly make it more potent.

I created several large iron trays, and poured the water into a thin layer to make sure it evaporated faster, and kept that tray away from the flame, barely bringing it to a low simmer. Not enough to boil it, but enough to hasten the evaporation.

That was the closest to the flame. The others, I kept farther away to better understand the impact of the heat.

And, once I set those trays and left them to evaporate, I went upstairs once more, dropping another thousand shells and watching the operation a bit. Then, I went to the second floor, and five minutes later, I returned with a broken cart and picked a new one. Maybe it was slightly paranoid, but considering Thomas killed more than a hundred people in his latest attempt, I didn't consider it below him.

It was better to keep any potential spy guessing.

Once I returned to the fourth floor, I started experimenting with the water in the trays. The damage to the plants confirmed that evaporation was a good way to increase the poisoning effect on the plants, but heat somewhat damaged the process.

"How fun," I muttered. The effect of the heat made the creation of poison water difficult, but not impossible. Luckily, I had the steam engine to provide constant motion.

First, I created a huge water reservoir, made of entirely metal, with an adjustable sieve at the bottom. It had a channel, which constantly dripped water into a centrifuge powered by the steam engine, which rotated at around 400 RPM. As the centrifuge rotated, the particles moved toward the edge, but spilled only when the water amount had turned too much.

Meanwhile, there was a small gap at the center, which allowed the purer — poisonous — water to flow to another channel, which led to multiple, layered trays that collected the water. Each layer, except the bottom one, had a Pythagorean cup setup, meaning water had been collected slowly in a layer, and once it reached a certain height, it started to drain until it reached the minimum level.

Not too different from using a hose to drain a gas tank.

And, to complete, there was a ventilator attached to the steam engine, creating wind to enhance evaporation speed.

As each layer evaporated, the ones at the lower layers started to gather denser poison. That way, the process was sustainable without constant supervision — other than cleaning the sieves occasionally — giving me a selection of poisons of different concentrations to choose from, which I could use to improve my Nurture skills.

Since I needed to improve my Nurture ability quickly, I built the tray system larger than necessary. It consumed almost fifteen metric tons of corrosion-resistant iron, with each tray spreading almost a thousand square feet, which was the size of a decent three-bedroom apartment. Connected to each tray, there was a reservoir that didn't drain even when the suction setup was triggered, allowing me to get a sample from whichever layer I wished.

As I worked on adding more trays to the bottom, the first layer reservoir and the tray had been filled, and the tray section had been drained thrice, while the second reservoir had collected a decent amount.

Unfortunately, the evaporation process required some time, meaning I needed to work more.

"Let's try it," I said even as I scooped a small bucket from both the first and the second reservoirs. And, I used the first one on a mature tree.

[-48 Health]

[Nurture (Rare) 97 - 99]

"Not bad, but it could be better," I said even as I took a cutting from the tree, and grew a new one.

[-123 Health]

[Nurture (Rare) 99 - 106]

"Much better," I said even as I repeated it a few times. The poison from the first layer lost its effectiveness at just above a hundred and thirty attempts, while the second one kept me going a bit more. When I finished, my Nurture skills had developed significantly.

[Nurture (Rare) - 152]

It was promising. Once the fourth reservoir had enough poisoned water, I should be able to complete it, but I had no intention of breaking the setup once I completed it. After all, since each layer constantly evaporated, it would take longer and longer for the lower layers to be filled with water.

Without a doubt, the Epic variant of the skill would require even more potent poisons.

There was another benefit of keeping the setup. It could be used to rapidly increase the skills of the other Farmers, which might actually unlock better variants if we pushed it far enough. Maria had already confirmed that class upgrades were not a secret.

However, I didn't know if there was any difference between doing that through skill evolution and skill stones. Testing that with a few volunteers might actually help. However, at this point, I couldn't do that with the eight thousand farmers currently working. I had already rocked the boat enough for the moment.

Once again, just because it was an open secret for the powerful families didn't mean I could reveal it without putting a target on my back.

Well, another target.

"Patience," I said even as I moved toward my mining pit to crash some more ore, refilling the ore feeder for the blast furnace. However, I noticed that the deposit was running out.

It was good that I was planning to start an expedition once I improved my Nurture skills to the limit. Since I was going to create a path with the trees, adding some rails and a small, steam-powered engine to bring back the ore wouldn't be too difficult.