How to Split a Hive: 3 Methods, Step by Step

A split is a deliberate, managed swarm. You take brood, bees, and stores out of one colony and set them up as a second colony, and one of the two halves ends up with a new queen. Everything a swarm does on its own timetable and in a neighbor's tree, you are doing on your timetable and in your own box. That is the entire concept, and once it clicks, most of beekeeping's spring anxiety goes away.

What it buys you, in order of value. A free colony. Packages run well north of $150 and nucs more, while a split costs a box and some patience. Swarm prevention. A colony that just lost half its bees and most of its brood is not going anywhere; see the swarm prevention guide for where splitting sits in the sequence of levers. Free queens. A walk-away split produces a queen raised by your own bees from your own stock, which is also the entry point to real queen rearing. A brood break. The queenless half has no open brood for roughly three weeks, and varroa reproduces only in brood, which makes a split a genuine mite-management tool alongside the treatments in when to treat for varroa. Insurance. Two colonies fail less often than one, and the spare queen in a split is the fix for the deadout you find in March.

What it costs you: honey. A colony split in May will not produce a surplus crop that year, or will produce far less. That is the trade, and it is why sideliners split some hives and leave others alone. If your goal for the season is honey in jars, read how much honey per hive first and split your second-best colony, not your best one.

Five conditions, and you want all five. Missing one is a reason to wait a week, not a reason to improvise.

The single most common first-split mistake is splitting a colony that merely looks busy. Traffic at the entrance is not brood volume. Pull frames and count: if the whole colony holds four frames of brood, splitting it makes two weak colonies that both limp, and limping colonies get robbed, get chilled, and get overrun by small hive beetles. Half of a strong hive is a colony. Half of an average hive is two problems.

The walk-away split is the purest version: divide the colony, make sure the queenless half has eggs, and walk away. The bees choose several young larvae, feed them royal jelly, build queen cells, and in about four weeks one of those queens is laying. It costs nothing but time, and the queen it produces is adapted to your yard because she came from bees that survived there.

The mechanics are simple. Find the queen, or at minimum know which box she is in, and split the frames so the queenless side gets at least one frame carrying eggs and larvae younger than about three days, because that is the age window from which bees can still raise a queen (Wikipedia: queen bee). Give both halves brood, stores, and bees per the frame recipe below. Close them up, feed both, and do not open the queenless half for two weeks except to confirm capped queen cells at day 7 or 8.

The costs of the free method are real: three weeks with no new brood in the queenless half, which is a population dip right when you want growth, plus mating risk. Some virgins never come back, and a queen lost on a mating flight costs you another month if you have to start over. Plan for roughly one failure in five and keep a frame of eggs available from the parent colony as your reset button. If you want to steer the genetics instead of accepting whatever cells the bees build, a grafting setup like the APlayfulBee queen rearing kit ($19.71) is the cheap way into cell-building on purpose rather than by accident.

Buy a mated, laying queen and the timeline collapses. Instead of 28 days of waiting and a coin flip on mating, you have brood inside a week and a colony that never stops growing. In a northern season, or any season where the split needs to be a full-strength colony by September, this is usually the right call even though it costs $35 to $60.

The sequence matters more than with a walk-away. Make the split first and leave the new half queenless for a few hours, ideally 4 to 24, so the bees register the loss and stop treating an introduced queen as an intruder. Then place the caged queen candy-end accessible between two brood frames, screen facing into the seam so bees can contact her, and leave the cage in place for three to five days. Direct release is a good way to get a $45 queen balled and killed. Penn State's extension guidance on queen introduction is worth reading before your first one; our own step-by-step is in how to requeen a hive.

Two practical notes. Mark her, or buy her marked: a marked queen turns future inspections into five-second confirmations and tells you instantly if the colony superseded her. The benefitbee queen bee marking kit ($21.99) pays for itself the first time you need to know whether the queen in the box is the one you paid for. And order early: good queen producers sell out for spring by February, and the queens available in a panic in June are the ones nobody else wanted.

A nuc split is a smaller, gentler bite: pull five frames into a nucleus box instead of halving the hive. The parent colony barely notices, stays strong enough to make honey, and you still get an increase, a queen, and a brood break in the piece you removed. For most keepers with two or three hives this is the split to learn first, because the downside is capped: if the nuc fails, you lost five frames, not half a colony.

The classic five-frame recipe: two frames of brood with at least one carrying eggs and young larvae, two frames of honey and pollen, one drawn comb or foundation, and a generous shake of nurse bees from a third brood frame. Housing is a nucleus colony box, and it does not need to be fancy. The Mann Lake Pro Nuc ($38.81) is the standard for a reason: plastic, weatherproof, locking lid, ventilated, and light enough to carry loaded. If the nuc is going to sit out for a full season or overwinter, a wooden box like the BeeCastle 5-frame beeswax-coated nuc ($75.71) handles weather and insulation better; that decision is the whole subject of how to overwinter nucs.

Nuc splits also scale in a way half-splits do not. A very strong double-deep colony in May can donate two nucs and still make a crop. Pull them a week apart rather than on the same day, and pull from different boxes so you are not stripping the brood nest all at once. The gear side of nuc housing, including swarm-trap doubles, is compared in the nuc box roundup.

Splitting is not "half the frames each." It is a recipe, and the queenless half gets the heavier share because it is about to lose its foragers and has to raise a queen from scratch. Here is the working allocation for a strong ten-frame deep, split into a full second box or a nuc.

Three rules that override the table. Know where the queen is. Find her and cage her briefly, or work a box at a time so you are certain. Splitting a colony and accidentally leaving both halves queenless, or both queenright, is the classic beginner outcome. Keep frames in order and orientation. Brood in the middle, stores outside, comb faces the way they were. Do not build a random shuffle; the bees have to cover it tonight. Never split brood off from its nurse bees. Brood without enough bees to blanket it chills, and chilled brood is dead brood the same night. When in doubt, shake in more bees.

Here is the thing that surprises everyone's first split. You divide the bees fifty-fifty, walk away, come back the next morning, and the half that moved somewhere new is half empty while the half that stayed put is packed. Nothing went wrong. Foragers navigate home to a remembered location, not to their family, and every field bee in the moved half flew out, worked, and came back to the entrance it has memorized.

The old rule of thumb: move a hive less than three feet or more than three miles. Inside three feet, returning foragers find the entrance anyway. Beyond three miles you are outside their normal foraging range, so they re-orient completely on their first flight out. In between, and that includes moving a split to the other end of the same yard, foragers stream back to the original spot. Nurse bees and young house bees, which have not yet flown, stay with the frames they are on, and that is exactly why the shake of extra nurse bees into the moved half matters so much.

You cannot beat forager fidelity, so you plan around it. Pick one or two of these, not all five.

1. Overload the moved half. Shake in the nurse bees from one or two extra brood frames before you close it up. Bees you shake in are disproportionately young, and young bees stay. This is the simplest fix and it is usually enough.

2. Move the parent, not the split. Leave the new queenless half on the original stand and move the queenright parent to the new spot. The parent has a laying queen and can afford the forager loss; the split gets the returning field force it desperately needs. This is the single best trick in the article and almost nobody does it on their first try.

3. The equalizing shuffle. Set both boxes side by side, both a foot or two off the original stand, so returning foragers split between two equally unfamiliar entrances. Bees are not precise enough to sort themselves perfectly, and the drift roughly balances.

4. Take the split out of the yard. Haul it more than three miles to a friend's yard for two weeks, then bring it home. Every bee re-orients at the new site and the split keeps its full complement. This is the most reliable method and the most work.

5. Confine briefly. Close the split with screen for 48 to 72 hours in a cool, shaded spot, with feed and ventilation, so the bees lose their fix on the old entrance. It works, and it also risks overheating and stressing a colony that is already in a fragile state. We use it least of the five.

Whichever you choose, reduce the split's entrance. A weak young colony with a wide-open door is a robbing target, and robbing kills more splits than failed matings do. A metal GJD entrance reducer and mouse guard set ($18.99) does the job for a couple of dollars a hive; the full field of reducers and screens is in robbing screens and entrance reducers.

Feed both halves. Not just the split, both. The parent colony lost brood and bees too, and a split made during a spotty flow can leave either side short within a week of bad weather. The standard is 1:1 syrup by weight, which mimics nectar and stimulates comb building and brood rearing, and you keep feeding until the colony is drawing comb and packing its own nectar, then stop. The full method, mixing without scorching, when to stop, and the never-feed-with-supers-on rule, is in how to feed bees sugar syrup.

Feeder choice matters more for splits than for established colonies, because a small colony cannot defend itself. Feed inside the hive. A hive-top feeder like the KAGZ 1.5-gallon hive top feeder ($37.99) holds enough syrup that a young colony can go a week between refills, and the bees access it without ever leaving the hive, which means no syrup smell at the entrance and no robbing invitation. Entrance feeders are convenient and are the worst option for a weak split for exactly that reason. The whole comparison lives in the feeder guide.

Feed pollen too if your split is going into a pollen gap. A queenless split raising its own queen does not need much protein for two weeks, but the moment she starts laying, the colony needs pollen to feed larvae. A patty on the top bars covers a dearth; the options are ranked in pollen patties and substitutes. A feeding stimulant with essential oils, added to syrup, encourages uptake in cool weather when bees are reluctant to break cluster for cold sugar water.

The most common way to ruin a walk-away split is to keep opening it. Here is the actual clock, and the discipline it demands.

Bad weather stretches this. A week of rain during the mating window pushes everything back, and a queen mated on day 30 is normal, not a failure. Give a slow split until about day 32 before acting. If day 32 arrives with no eggs and no queen, add a frame of open brood with eggs from another colony: if the bees build queen cells on it, they were queenless and you have restarted the clock, and if they do not, there is a virgin in there you simply have not seen. That test is worth more than an hour of hunting.

The other tell: laying workers. If you find multiple eggs per cell, eggs stuck to cell walls rather than standing in the center, and scattered drone brood in worker cells, the colony went queenless long enough that workers began laying, and a purchased queen will usually be killed on introduction. The recovery playbook is in how to requeen a hive, and in fall the honest answer is often the newspaper combine in how to combine two hives.

Swarming is a colony reproducing, and it happens when the brood nest is congested, the flow is on, and the colony has the population to spare (Wikipedia: swarming). A split removes all three conditions at once, which is why it is the most reliable swarm control anybody has: no other lever is as decisive, and unlike cutting queen cells, it works on the cause rather than the symptom.

Timing separates preventive splits from emergency ones. Preventive: made before queen cells appear, when the colony is simply strong and filling up. You control the frames, the queen, and the timing, and both halves come out clean. Emergency: made after you find charged swarm cells, which means the colony has already decided. Here you split anyway, but the recipe changes: leave the cells with the queenless half, take the old queen out with several frames into a nuc, and let the parent finish the cells it started. That is very close to what the colony was going to do, minus the part where the bees end up in a tree.

What does not work: cutting queen cells and calling it solved. A crowded colony rebuilds cells within days, and by the third round the beekeeper is losing a race the bees will win. Cutting cells without fixing crowding also risks removing the colony's only cell after the old queen has already left with a swarm you did not see. The full ordering of levers, space first, then checkerboarding, then splits, then requeening, is in how to prevent swarming, and if a swarm does get away, how to catch a swarm covers the recovery.

1. No eggs in the queenless half. The colony cannot raise a queen and dwindles for a month before you notice. Confirm eggs on a frame before you close the box, every single time.

2. Both halves got queens, or neither did. Almost always a queen you did not locate. Work slowly, box by box, and if you truly cannot find her, split so that both halves get eggs. Whichever half is queenless will raise one, and you can sort out the extra later.

3. Too weak to cover brood. A cold night after an over-thin split chills brood across the frame edges. Split strong colonies only, and shake in more nurse bees than you think you need.

4. Robbed out. A small colony with a wide entrance during a dearth attracts every forager in the yard. Reduce the entrance to an inch, feed inside the hive, and never spill syrup near a split.

5. Starved during a weather hole. A split with one thin frame of stores runs out during ten days of rain. Give each half real capped honey, not just syrup, and check the heft weekly.

6. Mating failure. Cold, wet weather during days 17 to 24, or no drones in the area. This is the argument for a purchased queen in a short season, and for keeping a spare frame of eggs available.

7. Opened too often. Every inspection during the virgin's mating window risks losing or balling her. Confirm capped cells at day 8, then leave it alone until day 28. The general discipline of not over-inspecting is covered in how to do a hive inspection.

Week 1. Check syrup levels on both halves and confirm the entrance reducers are in. Do not open the queenless half beyond a lid crack. Watch entrances: pollen going in on the parent means the queen is laying; the split will not show much yet.

Week 2. Confirm capped queen cells in the queenless half if you have not already, or confirm the purchased queen is out of her cage and the colony is calm. Refill feed. Note stores.

Weeks 3 and 4. Hands off the queenless half. Inspect the parent normally, and watch it for renewed swarm signals, because a strong colony can rebuild crowding surprisingly fast after a split during a good flow. Add a super to the parent if it needs one; a MayBee medium super box ($44.99) is the standard unit for giving space in a hurry.

Week 5 (day 28 to 32). The verdict inspection on the split. Eggs and a tight young brood pattern mean success; mark the queen while she is easy to find in a five-frame box. No eggs means run the frame-of-eggs test.

Week 6 and beyond. A successful nuc split typically fills five frames in three to five weeks of decent flow and needs promoting to a full box or a second nuc story. From here the split's whole first year is the same job as any young colony: build comb, monitor mites, pack stores. If it is going into winter as a nuc rather than a full hive, plan that now, not in October, with how to overwinter nucs.