Do this before you clean up

Work in daylight, take the frames out one at a time, and photograph anything you cannot interpret. If you keep notes per hive, write the autopsy against that hive while you are looking at it, because by next autumn you will remember that you lost a colony and not what it looked like inside.

Timing depends on which half of the world you are in. In the northern hemisphere this is a late February to April job. In the southern hemisphere it lands around August and September. Either way it is the end of your winter, not a month on a calendar.

The one-minute test that settles most cases

Take a frame that held brood, hold it so light falls into the empty cells, and look at the upper inside wall of each cell. Mite droppings appear as small white crystalline specks. A few scattered specks mean little. Cells heavily speckled, across many cells, mean mites were reproducing in there in numbers.

This is the most informative thing you can do to a deadout, it costs a minute, and it is the test that separates the most common cause of winter loss from the two things people usually blame instead.

The signatures

Varroa and the viruses it carries

The most common cause of winter loss by a wide margin. What you find is a very small cluster or almost no bees at all, and stores still in the box. Sometimes a handful of bees remain, sometimes none. You may see deformed wings on the bees that are left, a patchy last round of brood, and speckled cells.

The confusing part is that it does not look like a disaster. It looks like the bees left. They did not: the colony dwindled through late autumn and early winter as mite-damaged bees failed to live out their winter lifespan, and the last few hundred could not hold a cluster. Full frames and no bees is the signature.

Starvation

Bees head-first into empty cells, often with a tight ball of dead bees around them. Two quite different versions:

  • Absolute starvation. No stores anywhere. The colony went into winter light, or ate more than it should have because it was too warm for too long.
  • Isolation starvation. Honey still in the box, but not where the cluster was. In sustained cold a cluster cannot break formation to cross empty comb, so it starves a few frames from full ones. This is a placement problem rather than a quantity problem, and the fix is making sure stores sit above the cluster rather than beside it.

Damp

Mould on comb and hive walls, water on the floor, staining under the crown board, a small wet cluster. Distinguish it from the mould that grows on any deadout after the fact: what points at damp as a cause is water in places water should not have been while the colony was alive.

Queen failure

No brood at all going into winter, or a last patch of drone brood in worker cells, or several eggs in one cell from laying workers. The colony dwindles because nothing replaced the bees that died. Often traceable to a supersedure or a mating failure the previous autumn.

Robbing

Torn and ragged cappings, wax flakes across the floor, comb chewed rather than emptied. Robbing is usually the finish rather than the cause: a colony already weakened gets stripped. Worth noting because it tells you the colony was weak first, and the question is still why.

Mice

Unmistakable. Shredded nest material, droppings, and comb chewed away in tunnels. Mice rarely kill a strong colony outright, but they destroy comb and add stress a marginal colony does not need.

Things that look like causes and are not

  • Wax moth. A consequence. Wax moth moves in once there are not enough bees to keep it out, so it tells you the colony was already failing.
  • Mould everywhere in spring. Also usually a consequence. A dead hive goes mouldy whatever killed it.
  • "It froze." Almost never the primary cause. A dry cluster with enough bees and enough food handles remarkable cold. Cold finishes colonies that were already too small, too wet or too hungry, and the useful question is which.
  • Dead bees on the floor. Normal. Bees die all winter and the survivors cannot always carry them out. A layer of dead bees on the floor is not by itself a diagnosis.

Stop if you suspect foulbrood

Sunken, greasy or perforated cappings, discoloured larval remains, a ropy thread when a suspect cell is probed with a matchstick, or hard scale stuck to the cell wall. American foulbrood is notifiable in many countries and the equipment must not be reused.

If any of that is present, close the hive up, keep it away from other colonies and other people's bees, and contact your bee inspector or the relevant national body. That is the one finding where the right next step is a phone call rather than an article.

What to do with the equipment

With foulbrood ruled out, most deadout equipment is reusable and drawn comb is genuinely valuable. Scrape the floor, brush out dead bees, and store frames somewhere cold, dry and protected from wax moth. Discard comb that is black, mouldy through, or full of crystallised stores you cannot reuse.

If mites were the answer, that comb is fine. Varroa dies with the colony and does not lurk in empty equipment.

Then report it

Several countries run annual colony loss surveys, and they only work because ordinary beekeepers fill them in, including the ones who would rather not admit to a loss. The Bee Informed Partnership runs the long-running survey in the United States, COLOSS coordinates monitoring internationally, and many national and regional associations run their own.

Losing colonies is normal enough that national figures routinely sit in the double digits. A loss is not a verdict on you. An unexamined loss, though, is a season's information thrown in a bin.

Sources and further reading

The month a deadout is found depends on your hemisphere, and the likely causes shift with climate. Anchor anything below to the end of your own winter.