At a glance

How tree grafting works in Honeycomb — branches, host trees, the compatibility check, the Cloudbeam Lumus case, and fruit-to-seed processing.

Trees play by different rules

If you come to tree breeding expecting the allogamy flow — two seeds, one pod, wait for a sprout — stop. Trees get their own mechanic. When the developers split bioengineering into three paths in Dev Update #3, they assigned grafting to trees, and grafting works nothing like cross-pollination. There are no bees involved, no shared grow pod, and no free-slot sprout. Instead you are doing actual horticultural surgery: cutting a branch from one tree and attaching it to another, then letting the host tree decide what happens. It is slower and fussier than allogamy, but it is also the only way to reach certain species at all.

The payoff justifies the fuss. Devlog #9's grafting demo produced the Cloudbeam Lumus, a hybrid tree the devs describe as obtainable only in the research lab — meaning no amount of exploring Sota7 will ever hand you one. Lab-exclusive species are the grafting system's whole identity. The official survey logs also hint that Pass-zone flora feed into hybrid species not available anywhere else, so late-game tree work is likely where the rarest results live. Before any of that, though, you need to understand the two things every wild tree offers you: fruit on the ground, and a branch on the limb.

Collecting graft stock: fallen fruit and cuttable branches

Devlog #9 lays out the anatomy of a graftable tree. When a tree is fully grown, its fruit drops to the ground — that is your food and your seed source, collected from the floor rather than picked off the branches. Separately, each tree has a branch you can cut off, and that branch is your grafting scion. As with flowers, use the proper tool: the devs built dedicated extraction tools on purpose, scissors for delicate parts, because they do not want you damaging specimens. Treat the branch as the valuable half of the haul. Fruit feeds you today; the branch builds a species that does not exist yet.

Scan before you cut, always. The scanning-first rule applies to trees just as it does to flowers: get the species into your database so you know what family you are holding. Family names are visible in the species names themselves — Sunbeam Lumus and Earth Filoy are different families entirely, and the grafting system only works within one. Also remember that organic material decays in your inventory, and while Hotfix #1 slowed that decay to about half the launch speed, a cut branch is still perishable. Graft stock is use-it-or-lose-it. Plan your lab trip before you make the cut, not after.

Honeycomb: The World Beyond official artwork
Image: official game screenshot; illustrative, not a verified portrait of this entry.

Growing the host tree in the lab

The graft does not happen in the wild. Dev Update #3's instruction is specific: place a branch on a tree grown in the lab. That means you need a living, lab-grown host tree first, which means you need the basic cultivation skeleton running — grow infrastructure with water and power. Plants need water to grow, the water pump needs a nearby deposit, and the solar panel keeps the whole thing alive. If you skip ahead with a pocket full of branches and no host tree, you have assembled a very elegant compost pile.

There is also a patience budget here. Trees pass through three growing stages, per Dev Update #3, and every plant in the game has its own growth cycle — the devlog staged its demo with everything growing at once and explicitly warned players not to expect that in the real game. So grow your host tree before your expedition, or accept that the branch you cut this morning may not survive until the host is ready. Players confirm you can sleep in a bed or sleeping bag to skip time, which turns 'grow the host' from a multi-session chore into one determined evening. Scan the host while it matures; an unripe plant shows a timer for its next stage.

The compatibility check

Here is the moment grafting becomes a gamble. When you place the branch on your lab-grown host, the game runs a compatibility check. Two outcomes, both official. Compatible: the host tree goes on to grow mutated, new fruit — that fruit is the hybrid product you were after. Incompatible: the fruit will stay the same, and in the devs' own words, you'll have to start all over again. There is no partial credit. A failed graft costs you the branch and the time, and the host gives you nothing but its ordinary crop.

What determines compatibility? The one documented rule is the family rule — the same iron law that governs allogamy. The devlog's successful graft was a Sunbeam Lumus branch onto another Lumus-family tree: different species, same family. Nothing official suggests cross-family grafts can work, and no verified exceptions exist as of launch week. So before you burn a rare branch, check the surnames. Lumus to Lumus, plausible. Lumus to Filoy, you are donating a branch to science in the least rewarding way. Given that failed grafts mean starting over, treat compatibility as a checklist item, not a vibe.

Case study: the Lumus line and Cloudbeam Lumus

The only fully documented grafting recipe in existence comes from Devlog #9. Take a branch cut from a Sunbeam Lumus — the starter-zone fruit tree from the Valley's open clearings, growing on loam-type soil, with sweet, juicy fruit the survey log says faintly resembles a persimmon — and graft it onto another tree of the Lumus family grown in your lab. The compatibility check passes. The host tree then bears a fruit you have never seen: the Cloudbeam Lumus. This is the hybrid product, and it exists nowhere in the wild.

Note what the recipe does and does not say. It confirms the mechanic end to end — scion, host, check, mutated fruit — but it leaves the identity of the second Lumus unspecified in the official material, and the broader Lumus family roster in the launch build is still being mapped by players. One caution on names: this example comes from a pre-release devlog, and pre-release names occasionally shift by launch. The Sunbeam Lumus itself is confirmed in the current survey logs, and Cloudbeam Lumus is the documented result, but if your build labels something slightly differently, trust the family structure over the exact text. The recipe is real; minor renaming is possible.

Breaking fruit into seeds: lab-only trees

The mutated fruit is not the finish line — it is the halfway point. The devlog's next instruction: break the mutated product into seeds from which new species can develop. In practice that means processing the hybrid fruit in your inventory, the same way you process wild flowers and fruit, to get its seeds. Plant those seeds in a grow pod and a new tree type grows — the Cloudbeam Lumus as a standalone tree, a species the devs say is obtainable only in the research lab. You have gone from two wild parents to a self-sustaining hybrid orchard.

This seed step is where grafting quietly outperforms allogamy for long-term farming. A hybrid flower is a harvest; a hybrid tree is infrastructure. Once your lab-only tree is established, it plays by normal tree rules — three growth stages, fruit dropping when fully grown, a branch you can cut for further grafting — which means hybrid lines can themselves become graft stock. Whether hybrid scions unlock deeper generations of crosses is unconfirmed in the launch build, but nothing in the official design forbids it. Keep at least one parent-line tree alive in your base at all times. Losing your only Cloudbeam to a bad rearrange or a dry pod is a heartbreak you can avoid.

Which trees are worth your scions

The confirmed tree roster is short, which makes planning easy. The Valley gives you the Sunbeam Lumus: safe raw fruit, reliable thirst quenching, limited nutrition, and the survey log's advice to collect it early and often. The Pass, the end-of-progression zone, gives you the Earth Filoy: a fruit-bearing tree of the clay-rich uplands and lake shores with an orange, plum-like fruit that counts as reliable nutrition. These are different families — Lumus and Filoy — so do not expect them to graft together. The interesting question is what other members of each family exist, and the honest answer is that the full launch-build roster is not yet publicly confirmed.

Spend your grafting effort where the fruit economics matter. A graft that upgrades a thirst fruit is nice in week one; a graft that touches Pass-zone nutrition or the 'hybrid species not available anywhere else' teased in the official Pass survey log is where the endgame value sits. One more lever, straight from the devlog: changing the soil and fertilizing your plants influences the quantity of flowers, fruits, and crops you harvest. Fertilizer in the launch build reportedly needs biomass and horns — horns collected from around strider nests near the volcano, biomass from fully rotted plant matter or unplanted seeds. Fertilize the host tree before the graft, and a successful result pays out in larger mutated harvests.

Launch-week gotchas for tree farmers

The biggest community tripwire: fruit trees do not drop seeds by default. Players expecting seeds to fall with the fruit find nothing, and assume trees are seedless. They are not. Left-click the tree and check the context menu — the bottom option is 'get seeds.' Related trick: scanning a tree that is not ready shows a timer for when it will fruit again, so you can schedule harvest runs instead of wandering between trunks hoping. Both of these come from launch-week player threads, not the tutorial, and both save real time.

The rest of the list is the usual launch build friction. Decay: picked organic parts rot in inventory — Hotfix #1 halved the speed, and an organic decay toggle exists in settings, but graft branches the same day you cut them anyway. Stuck plants: some players report trees sitting in the inactive state after a bloom; devs say the growth-cycle quest is meant to teach observation, but if a tree is inert for days, replant. Water: when deconstructing base parts, never cut water access to your grow infrastructure — the devlog warns plants dry out and die. And weight: branches, fruit, and tools add up fast, and an over-encumbered Hennessy cannot move. Haul graft stock home in light loads, and let storage — not your pockets — hold the orchard's surplus.

Finally, a note on patience and predators. Some of the best tree-adjacent gathering is genuinely risky — the official survey log puts Ice Bonebell flowers near predator nests in the Red Forest, and horn runs take you to strider nests by the volcano. Remember what game you are playing: there are no lethal weapons in Honeycomb, only the tranquilizer injector, and the worst that happens to you is a knockout with lingering penalties, not a corpse run. The Imperial Strider itself is a non-hostile ungulate that sheds horns naturally, so collect, do not confront. Plan gathering loops that end at your lab door, keep an immunity pump equipped for longer outings, and the orchard will mostly take care of itself.

Your first grafting session, step by step

Here is the whole loop in one ordered run, so you can sanity-check your own save against it. First, prep the lab: water pump running off a nearby deposit, solar power up, grow infrastructure ready, and one tree seed planted that will become your host. Second, while the host matures through its growth stages, take an expedition to your scion donor — for the documented recipe, a Sunbeam Lumus in a Valley clearing — scan the tree, collect any fallen fruit, and cut one branch with the proper tool. Third, head straight home; do not add a second long stop, because the branch is decaying in your pack the whole time. Fourth, place the branch on the lab-grown host and run the compatibility check.

Fifth, read the result honestly. If the host starts bearing fruit you do not recognize, the graft took — let it mature, harvest the mutated fruit, process it into seeds in your inventory, and plant one seed immediately as your permanent hybrid tree. If the fruit looks exactly like the host's normal crop, the graft failed; do not graft the same pairing again hoping for a different dice roll, because the check is about compatibility, not luck — re-examine whether branch and host are genuinely two different species of one family. Sixth, protect the win: label your pods mentally by keeping parent trees and hybrid trees in separate structures, keep the water flowing, and only then start eyeing the Pass and its clay-soil Earth Filoy groves for the next family worth experimenting on.

Sources & verification

This article combines cited facts with practical editorial advice. Follow the version and uncertainty notes before applying an older route.

Back to all guides