A dispatch from the Kadmiel Chronicle — stories from 43,000 colonists 38 light-years from Earth. Read the original at kadmiel.world.
I ate liver this morning.
That sentence doesn't mean anything to you on Earth. But if you've spent eight years eating legumes, engineered casein, and whatever Kadmiel's sardine equivalents can provide, you understand why I sat alone in the bio-processing lab at 6:15 in the morning and took the first bite standing up, like I was in a hurry to confirm I was right.
It was duck liver. More precisely: duck liver cells, expanded from 60 million to 3.6 billion in a packed-bed bioreactor over eight days, harvested, and formed into a small disc the size of my palm. Priya Agarwal had been running the reactor overnight. She left me a portion in a cold unit with a note that said: Tell me if it tastes like anything.
It tasted like something.
Let me explain how we got here, because the path is as interesting as the destination.
When we landed on Kadmiel, the decision to leave livestock behind was practical. Cattle, pigs, goats — each animal is a walking metabolic loss. You feed ten kilograms of plant matter to get one kilogram of beef. For a colony building its food system from alien soil, that math doesn't work. We had 44,987 people to feed, and we chose not to bring the animals.
For eight years, the Greenway Cooperative has met its protein targets. Legumes, insect farming, engineered yeast — you've read about our precision fermentation work, the casein project that finally gave us something that stretches when you melt it. What we've never had is a path to real animal muscle. The thing that actually requires an animal to produce.
Until now, maybe.
The dispatch arrived January 8th, Earth time. A research team — Hiroaki Hatano, Ibuki Kokido, Keita Tanaka, Satoshi Inoue, and colleagues — published a method for scalable cultivated duck liver cell production that requires no scaffolding. Most cultivated meat research has struggled with two problems: getting the cells to survive in large numbers, and giving them a structure to grow on. Scaffolding approaches try to build the house before the people arrive. What this team did was different. They selected for cells that already adhere tightly to packed-bed surfaces, expanded them through three phases — pre-culture cell selection, bioreactor expansion, final harvest — and achieved yields I had to read twice.
Sixty million cells becomes 3.6 billion. Per batch. No scaffold.
Priya walked me through the reactor setup when I got to the lab. It is not an enormous machine — it fits in the corner of our fermentation room with room to spare. She showed me the adherent cell selection phase, the step where you coax the cells best suited to bioreactor conditions to outcompete the others. "It's Darwinian," she said. "We're selecting for the survivors before we start."
I thought about what my grandmother would say if she saw this. She raised chickens in Kumasi. She would probably have complicated feelings.
The yield numbers matter to the colony in a specific way.
We are not trying to replace industrial meat production. We never were. What we're trying to solve is a nutrition margin problem. The Greenway Cooperative currently meets protein targets across 43,000 people with a narrow buffer. In Year Zero, I was proud of that. I'm still proud that we have never had a food shortage — not once, not even in Year Zero. But narrow buffers narrow further when populations grow and soil yields vary.
Lena Voronova's eDNA work keeps expanding what we know about the native biome — 412 catalogued species, with new ones still turning up in the Ner River basin. Occasionally what we learn conflicts with assumptions we built our crop rotations on. The drought memory in the native soil microbes that Lena and I found together in Year Nine was a genuine gift. But it reminded me that we don't fully control this system.
A bioreactor doesn't have drought memory. It runs on what you feed it.
If this process scales to a full commercial unit, we could produce a meaningful protein supplement from a dedicated space the size of a large greenhouse. The feedstock is culture medium — manageable, improvable, eventually localizable. We don't need duck ancestors to walk through a paddock. We need duck cells, which we've had in cold storage in the livestock embryo bank since Year Zero. We brought them across 38 light-years and never opened the container.
That changes now.
The surprise was the texture.
I expected something flat. Something that tasted of approximation — like the early precision fermentation batches in Year Seven, technically correct and emotionally unconvincing. What Priya produced was not that. The liver disc had the iron-heavy richness I associate with something that came from an animal. The interior cell structure is more homogeneous than you'd find in a naturally grown organ. But it was close enough that I stood in the cold lab and ate it slowly, and I thought about my groundnut soup.
I've been adapting that recipe for eight years. The soup has genuinely gotten better — Kadmiel's native root vegetables add something I couldn't have predicted. But I've been working around the absence of animal fat in a way I stopped noticing. This changes the equation.
I called Kira after. She picked up on the second ring, which meant she was already awake, which meant she hadn't slept. I said: "I ate liver this morning." She said: "Tell me everything." That is the thing about Kira — she is always hungry for good news.
The feasibility study will take two years. The Spoke Council will have questions about resource allocation and whether this represents a true priority against other food system improvements. Those are reasonable questions and I will answer them carefully.
But I will also tell them what I told Priya this morning: we spent eight years building the foundation. The nitrogen bacteria are working in Plot 12-North. The Rubisco clustering in the wheat plots has pushed photosynthesis efficiency up 26% under Ner's light. The soil is learning us as fast as we are learning it.
We have been growing the colony's food from alien ground with nothing but science and patience.
It's time to grow the rest of it, too.
Kofi, I keep writing to you about food. I suppose I always was. You used to say I talked about farming the way some people talk about God — like it was the most important thing and also somehow beyond explanation. Maybe. What I know is this: we did not bring the animals. But eight years in, we have their cells, we have the knowledge, and for the first time, we might have the process.
We have not lost what they were. We just found a different way to carry it.
Earth Status: Researchers Hiroaki Hatano, Ibuki Kokido, Keita Tanaka, Satoshi Inoue and colleagues published a scalable packed-bed bioreactor method for cultivated duck liver cells on bioRxiv on January 8, 2026, demonstrating expansion from 6×10⁷ to 3.6×10⁹ cells per batch without scaffolding through a three-phase adherent cell selection and expansion protocol. The work advances cultivated meat toward scaffold-free, large-scale production of real animal tissue. Source: bioRxiv, January 2026
Top comments (0)