Close This Window

Posted 18 May 09

Where do Honeybees go in the Winter?

Eric McArthur


The article below was printed in the September 1982 Scottish Beekeeper magazine.  It was written as a result of observations made on some 60 - 80 colonies over wintering over a prolonged period of time.  My method of management then, involved entering colonies even in the dead of winter when necessary.  During these manipulations I came to recognise that there was always patches of brood of varying size in the colonies at all time during the winter period – as had Professor Friedrich Ruttner, when the Bee Institute at Oberursel in Germany were investigating the effectiveness of a variety of anti Varroa treatments.

When the article in question was printed it attracted a lot of criticism especially from Bernhard Möbus whose rebuttal was printed in the May 1983 S.B magazine. The deliberations in the article which I wrote were at the time purely academic all my colonies at that time were acclimatised Scottish hybrids predominantly and black bees.

However it seems that the old adage about all things coming to those who patiently wait has some substance: Recently in quick succession two pieces of information arrived on my computer; the first was an e-mail from a correspondent of long standing, expressing astonishment at his observation with two separate colonies. This man works with acclimatised Scottish hybrids and Scandinavian Buckfast bees (which incidentally I do not approve of – but that is a separate issue!). The Scandinavian Buckfasts are famed for the yellow coloured abdomens, the acclimatised Scottish hybrids are quite un-striking in their colouring. My correspondent had re-queened a colony of black bees using Scandinavian Buckfast queens at the end of 2008.

I have been in hot water many times in my beekeeping life but there are occasions such as these that make the slings and arrows of dogma eminently bearable.

Dear Eric
I thought that one very interesting aspect has come to light. I re-queened a couple of colonies very late last year. They were black and the queen was with an orange stripe. Conventional wisdom is that the autumn bees live over till now when they gradually die off on foraging trips. Well it was greatly to my surprise to find that in both colonies, there were very few black types to be seen. Just goes to show how much early brood rearing goes on which you don't recognise. Furthermore it shows the importance of the bees being able to collect late autumn pollen to store over and start the process before spring sources become available.
(Name and address supplied)


The second just as revealing and extremely pleasing result (for me anyway!) was published in Bee World, 48(2): 85 – 90 (2009), the organ of the International Bee Research Association – titled:-

“The Development of Honey Bee colonies on the Northern Great Plains of North America confined during the Winter”

A summary of the work is appended below –


Honey bee colonies confined to winter quarters were monitored from 5 December until 11 March to assess changes in sealed brood production, colony demographics and adult populations during winter confinement. Small amounts of sealed brood were observed to be continuously present in colonies throughout the winter. Enough brood was reared during the winter to produce a small but temporary increase in the adult bee population and to replace most of the adults that died during the winter. Requeened colonies produced slightly more sealed brood during winter than colonies that had retained their original package queen. Approximately 34 - 50% of the adults in colonies in March had been reared during the winter. In March, the average adult colony was composed of workers that ranged from a few days old to 192 days of age. Average spring adult populations were 13,274 ± 1,078 (range 5,000 to 20,746).


The accepted theory regarding the over-wintering of honeybees in temperate and colder climates, subscribes to the hypothesis that the hive population, especially that part of the population consisting of young bees produced at the end of the late summer/autumn from the last of the nectar and pollen sources, carries the honeybee colony through the winter dormant period.

I would question this popular concept, and submit the following arguments to substantiate my doubts.

1. In 1950 Anna Maurizio, an Austrian research worker, demonstrated that by feeding "typical summer bees" pollen, their life span increased from 24 days to 36 days. While by feeding "typical winter bees" whose life span had been demonstrated to be 36 days, the pollen had no bearing on the life span of these bees, which lived 36 days on their normal diet and 36 days even with supplemental pollen.

2. titlehough we beekeepers here in Scotland have good access to ling heather invariably and are able to have young populous colonies going into the winter, consider beekeepers in other parts of the country or countries which also "enjoy" temperate climates, where the colonies cease to rear brood in large quantities 2000 - l500 per day, relatively early in the season, perhaps even before the end of July in some regions. These colonies by the time the winter cluster is formed will comprise of primarily older bees, if we subscribe to popular belief.

3. Many prominent beekeepers, among them Farrar have made the observation that the colonies which perform best in the following spring/summer are those colonies which consume the most stores during the previous dormant period.

4. In recent years increasing numbers of observations on over-wintering colonies generally have confirmed that at no time in the winter cluster is there a total hiatus in brood rearing.

My own experience has confirmed the above observations in my own hives. Many years ago I discovered to my cost, the limitations of over­wintering bees in single brood chambers where. British Standard equipment is used. The Langstroth frame is much more suitable for single brood box over-wintering.

Over the years I have devised a system of management for B.S. single brood box over-wintering which titlehough is not perfect allows me to "get away" with the single brood box.

This system consists of maintaining a close watch on colony progress during the winter period by raising the crown board and visually checking the disposition of the cluster. If I notice that the top of the cluster is up level with the frame tops in the brood chamber while the weather is still very cold, this tells me that the bees are hungry and have eaten their way through the normal "crown" of honey on the frames on which they are clustering .

I then remove the outside frames from the brood box (which will be solid with stores!), carefully prise the frames on which the bees are cluster­ing apart and insert at least two good frames of stores right into the heart of the cluster. I also give the colony a 1 kg. bag of sugar solidly crystallised, candy is perhaps a better bet in a case of this kind, or even a frame of honey laid fat on top of the frames.

During the manipulation described, which I have had occasion to carry out even in mid December in some years but certainly quite often during very late December and early January - in each case I have seen patches of brood on the combs!

5. In mild winters I have consistently noticed that the decrease in hive weight is much less than that over the same period where the winters have been accompanied with long periods of frost and little or no daytime sunshine.

6. I have observed over many years that colonies which at the return from the heather had expanded from a large 4-5 frame nucleus going to the heather to a colony which had filled the brood box with honey out to the outermost combs in the hive, and which at the first frost contracts to cover around 7-9 combs, will invariably over-winter better than a hive which went to the heather as a full strength honey gathering unit, which after the removal of the honey supers and even after the first frosts still covers every frame in the brood box which is of course as well filled with stores as the previously mentioned smaller colony.

My over-wintering system consists of an empty shallow crate, above the single brood box, this to house the contact feeder. Each colony is fed at least 10 lb of sugar in solution. The crown board is placed on the top of the super leaving an air space the depth of the shallow crate for ventilation. The hive entrance is 3/16” high and full width, winter and summer.

7. Making observations on hives where the queen has been lost during the early part of the winter, or where the queen becomes a drone layer during the winter period, I have consistently noticed that, regardless of the type of winter, whether hard or mild, by the third week of March, early April, that the populations of these hives have dwindled to virtually nothing - the loss of bees in such a colony once the bees begin to forage around the first week in March is quite dramatic.

8. Observing a queen right hive shows that the population falls quite rapidly during March/April, but at a particular level where the population is averaging perhaps 4-5 frames at say mid April - the population losses stabilise and the colony then begins to expand steadily until around the first week in May, when there is a "population explosion."

I'll try now to draw all the points made into some logical sequence.

If Anna Maurizio is right then a period of not much more than 8-9 weeks elapses between egg laid and dead winter bee, on average. Therefore in the case of bees returning from the heather, the last of these bees produced while the queen was still laying hard will have died or will die as soon as they move out of the hive an the "classic" winter cleansing flights, and at least by mid November.

If a colony enters the winter “overpopulated" this population will maintain itself during its life cycle by consuming valuable stores, where the moderate population will tend to consume less. I contend that the overpopulated hive depletes its provisions so severely, before these surplus bees disappear that by the time brood rearing begins the colony stores are well down, and from this observation I feel that if Farrar only got half the picture and should have qua1ified his statement with" Where this consumption of stores is converted to brood."

Regarding the increasing numbers of 'observations' confirming, the presence of brood in over-wintering colonies, at all times, I postulate that the full significance of this circumstance has been missed for years, and that this brood rearing, modest though it may be, is absolutely critical to the continued survival of the bee colony.

I would also go as far as to say that in reality by the end of March in most years where foraging begins at the beginning of March there are virtually no bees of the over-wintering population left in the hive. The rapid decay of population in queenless hives once foraging begins seems to confirm this idea.

Also the stabilisation of the population at around mid-April in a 'queen-right' hive and the level at which it stabilises is quite significant, since a colony covering 4/5 frames in early spring constitutes a number of bees ranging from 8,000 - 12,000.

I have made some rough calculations, and I would be grateful if the mathematicians among the readership could confirm or deny their accuracy! The results are based on an hypothetical queen in an  hypothetical over- wintering colony, laying eggs at the following hypothetical rates; thus:

1. From end of December to the end of February at the rate of 1000 eggs/week. Assuming that from egg laid to dead bee is 8 weeks I have used the 28 day month to simplify the calculation. Allow of course 3 weeks before the first brood begins to emerge.

2. From the beginning of March to the beginning of April the rate of lay increases to 2000 eggs/week.

3. From the beginning of April to the beginning of May the rate of lay increases to 3000 eggs/week.

Thereafter the rate will increase quite dramatically. Now I concede I am no great counter but surprisingly the number of new bees produced in the hive over this period is approximately 22,000 – 24,000 of which at the end of April there are 16,000 bees still alive. At the middle of April the population is 10,000 - which is not all that far from the expected "in viva" population at this time.

To produce 1000 new bees/week requires an area of approx. 60 sq. inches of comb, i.e. 20 sq. in. an each of three frames - based on 50 bees/sq. inch of comb, Considering both sides of the frame. This represents patches of brood approx. 4 in. x 5 in. on each of the three combs

I have noticed on many occasions much larger patches of brood, in late winter. Research has demonstrated that it costs about 1 lb of honey to produce 4,500 bees, thus to produce the number of bees in the hypothetical example, i.e. approx. 20,000 – 24,000 would cost the colony around 5 - 6 lbs of stores total plus of course the normal amount of honey to sustain the population of the hive, not a lot really!!

Eric McArthur


Close This Window