english,   español,   deutsche,   português,   русском,   română,   polsku,   беларускай,  

Natural Cell Size

And it's implications to beekeeping and Varroa mites

email address

"Everything works if you let it"
--James "Big Boy" Medlin


Standard foundation has been upsized

That upsizing has caused a bee that is 150% of it’s natural size

The fact that upsizing foundation makes a bigger bee and that we now have upsized is well documented by Baudoux, Pinchot, Gontarski, and most recently, McMullan and Brown.


Baudoux 1893

Made bees larger by using larger cells. Pinchot, Gontarski and others got the size up as large as 5.74 mm. But AI Root's first foundation was 5 cells to an inch which is 5.08 mm. Later he started making it 4.83 cells per inch. This is equivalent to 5.26 mm. (ABC XYZ of Bee Culture 1945 edition page 125-126.)


How much difference between natural and "normal"?

Keep in mind that "normal" foundation is 5.4 mm and natural cell is between 4.6 mm and 5.0 mm.

Volume of cells according to Baudoux:

Cell Width    Cell Volume
5.555 mm    301 mm³
5.375 mm    277 mm³
5.210 mm    256 mm³
5.060 mm    237 mm³
4.925 mm    222 mm³
4.805 mm    206 mm³
4.700 mm    192 mm³
From ABC XYZ of Bee Culture 1945 edition pg 126


Based on Baudoux’s actual measurements, this comes to a linear increase of 110% of the original size and a volume increase of 157% of the original size.

To help you visualize the difference, in human terms that’s like you naturally used to weigh 177 pounds and now you've been artificially manipulated so that you gained a hundred pounds and now you weigh 277 pounds. Or you naturally weigh 280 and now you've been artificially changed so you weigh 440 pounds.

Sevareid's Law

"The leading cause of problems is solutions."--Eric Sevareid


Foundation Today

Rite Cell Measurment Dadant 5.4mm Measured Pierco Medium Sheet 5.2mm Measured Pierco Deep Frame 5.25mm Measured Mann Lake PF120 Medium frame Measured Mann Lake PF120 Medium frame Measured Dadant 4.9mm Measured

Left to right: Rite Cell® 5.4 mm, Dadant normal brood 5.4 mm, Pierco medium sheet 5.2 mm, Pierco deep frame 5.25 mm, Mann Lake PF120 medium frame, Mann Lake PF100 deep frame, Dadant small cell 4.9 mm. NOTE: The Mann Lake PF100 and PF120 are not the same cell size as Mann Lake PF500 and PF520 frames which are 5.4mm.


Natural comb

4.7mm Comb 4.7mm Comb Measurement


Chart of Cell Sizes of Natural Comb and Common Foundation
Natural worker comb4.6 mm to 5.1 mm
Lusby4.8 to 4.9 mm average 4.83 mm
Dadant 4.9mm Small Cell4.9 mm
Honey Super Cell4.9 mm
Wax dipped PermaComb4.9 mm
Mann Lake PF100 & PF120 4.94 mm
19th century foundation5.05 mm
PermaComb5.05 mm
Dadant 5.1mm Small Cell5.1 mm
Pierco foundation5.2 mm
Pierco deep frames5.25 mm
Pierco medium frames5.35 mm
RiteCell5.4 mm
Standard worker foundation5.4 to 5.5mm
7/115.6 mm
HSC Medium Frames6.0 mm
Drone6.4 to 6.6 mm

Note: fully drawn plastic has thicker walls and is always .1mm larger at the mouth than the bottom and you have to allow for the thicker cell wall to come up with an equivalent. So the actual equivalent is pretty much the inside diameter of the mouth.


Why natural or smaller cells?
Varroa Life Cycle

Foundress enters the brood cell just before capping.

Lays one egg about every 30 hours.

First is male the rest are female.

Females have to reach maturity and mate to be viable and this takes 10.5 days from when the egg is laid.

Typical number of offspring in a worker cell with 21 day cycle (capped on day 9 and emerge 12 days later) is between one and two (1.5 +- 0.5) in a drone cell between three and four (3.5 +- 0.5). On small or natural cell the cycle is 19 days and the offspring are between none and one (0.5 +- 0.5)

During its time in the capped cell all of those Varroa, foundress and offspring, feed on the pupae weakening it and spreading viruses.

After emergence of the bee, the viable mites (the foundress mite and the one or two that made it to maturity and mated) go into their phoretic stage clinging to the bees and sucking their hemolymph like a tick, again spreading viruses and weakening the bees.


How do smaller cells help?
Male survivorship

Less male mites survive: Reproduction of Varroa destructor in South African honey bees: does cell space influence Varroa male survivorship? Stephen J. MARTIN*, Per KRYGER

Shortened Pupation
A model of the mite parasite, Varroa destructor, on honeybees (Apis mellifera) to investigate parameters important to mite population growth. D Wilkinson, , G.C Smith
More chewing out of varroa

Small cell and natural cell beekeepers have noted more chewing out of Varroa when regressing their bees.

Pre and Post Capping Times and Varroa

8 hours shorter capping time halves the number of Varroa infesting a brood cell.
8 hours shorter post capping time halves the number of offspring of a Varroa in the brood cell.


Accepted days for capping and Post Capping.(based on observing bees on 5.4 mm comb)
    Capped 9 days after egg laid
    Emerges 21 days after egg laid


Huber's Observations on Capping and Emergence on Natural Comb.

"The worm of workers passes three days in the egg, five in the vermicular state, and then the bees close up its cell with a wax covering. The worm now begins spinning its cocoon, in which operation thirty-six hours are consumed. In three days, it changes to a nymph, and passes six days in this form. It is only on the twentieth day of its existence, counting from the moment the egg is laid, that it attains the fly state."

François Huber 4 September 1791.

(note: this is a quote from the 1809 English translation and it is almost identical to the 1821 and 1841 English translations, all of which say "six days." However, I have since found the original French editions which say, in both the 1792 edition and the 1814 edition: "sept jours & demi" which should be translated 7 1/2 days. This makes it come to 20 days which is still one day short of 21 days)


Dzierzon's Observations on gestation on Natural Comb.

"When the young worker-bee has left the cell — which, reckoning from the egg, will be the case at the end of nineteen days, under favourable circumstances, but generally at the end of twenty to twenty-one days..."

--Jan Dzierzon, Rational Bee-Keeping, 1882 English edition, Pg 20


My Observations on Capping and Emergence on 4.95mm Comb.

I've observed on commercial Carniolan bees and commercial Italian bees a 24 hour shorter pre capping and 24 hour shorter post capping time on 4.95 mm cells in an observation hive.

My observations on 4.95 mm cell size
    Capped 8 days after laid
    Emerged 19 days after laid


What I've done to get natural comb


Things that affect cell size

o  Worker intention for the comb at the time it was drawn:
    o  Drone brood
    o  Worker brood
    o  Honey storage
o  The size of the bees drawing the comb
o  The spacing of the top bars


What is Regression?

Large bees, from large cells, cannot build natural sized cells. They build something in between. Most will build 5.1 mm worker brood cells.

The next brood cycle will build cells in the 4.9 mm range.

The only complication with converting back to Natural or Small cell is this need for regression.


How do I regress them?

To regress, cull out empty brood combs and let bees build what they want (or give them 4.9 mm foundation)

After they have raised brood on that, repeat the process. Keep culling out the larger combs.

How do you cull out the larger combs? Keep in mind it's normal procedure to steal honey from the bees. It's frames of brood that are our issue. The bees try to keep the brood nest together and have a maximum size in mind. If you keep feeding in empty frames in the center of the brood nest, put them between straight combs to get straight combs, they will fill these with comb and eggs. As they fill, you can add another frame. The brood nest expands because you keep spreading it out to put in the frames. When the large cell frames are too far from the center (usually the outside wall) or when they are contracting the brood nest in the Fall, they will fill them with honey after the brood emerges and then you can harvest them. You could also move the capped large cell brood above an excluder and wait for the bees to emerge and then pull the frame.

PLEASE do not confuse this issue of regression. I seem to get questions constantly asking whether to install a package on 5.4mm foundation first since they can't draw 4.9mm foundation well. If you want to get back to natural or small cell size, it is never to your advantage to use the already too large foundation they are already using. That is simply going nowhere at all. With a package, if you do so, you will have missed the opportunity to get a full step of regression. Dee Lusby's method is to do shakedowns (shake all the bees off of all the combs) onto 4.9mm foundation and then another shakedown onto 4.9mm to finish the main regression and then cull out the large comb until they have all 4.9mm in the brood nest. Shakedowns are the fastest method but also a stressful method and when you buy a package you already HAVE a shakedown. I would take advantage of it. If you intend to get back to natural size then STOP using large cell foundation all together. The main challenge is getting all the large cell comb OUT of the hive, so don't make that harder by putting more IN.

Another misconception seems to be that there are large losses in regressing. Dee Lusby went cold turkey, no treatments and only did shakedowns. She lost a lot of bees in the process. Many who tried the same also did. But this is not necessary.

First of all, there is no stress in letting them build their own comb. It's what they have always done. Second, it's not necessary to do shakedowns, it's just quicker. Third, you don't have to go cold turkey on treatments. You can monitor mites (and I would) until things are stable. Meanwhile you could use some non contaminating treatment IF the numbers get too high. I have seen no losses from Varroa from regressing in this manner and no increase in losses to stress related problems.


Observations on Natural Cell Size

First there is no one size of cells nor one size of worker brood cells in a hive. Huber's Observations on bigger drones from bigger cells was directly because of this and led to his experiments on cell size. Unfortunately, since he couldn't get foundation at all, let alone different sizes, these experiments only involved putting worker eggs in drone cells which, of course, failed. The bees draw a variety of cell sizes which create a variety of bee sizes. Perhaps these different subcastes serve the purposes of the hive with more diversity of abilities

The first "generation" of bees from a typical hive (artificially enlarged bees) usually builds about 5.1 mm cells for worker brood. This varies a lot, but typically this is the center of the brood nest. Some bees will go smaller faster.

The next generation of bees will build worker brood comb in the range of 4.9 mm to 5.1 mm with some smaller and some larger. The spacing, if left to these "regressed" bees is typically 32 mm or 1 ¼" in the center of the brood nest


Observations on Natural Frame Spacing

1 ¼" (32mm) spacing agrees with Huber's observations

"The leaf or book hive consists of twelve vertical frames... and their breadth fifteen lines (one line= 1/12 of an inch. 15 lines = 1 ¼"). It is necessary that this last measure should be accurate." François Huber 1806


Comb Width by Cell Size

According to Baudoux (note this is the thickness of the comb itself and not the spacing of the comb on centers)

Cell Size     Comb width
5.555 mm    22.60 mm
5.375 mm    22.20 mm
5.210 mm    21.80 mm
5.060 mm    21.40 mm
4.925 mm    21.00 mm
4.805 mm    20.60 mm
4.700 mm    20.20 mm
ABC XYZ of Bee Culture 1945 edition Pg 126


Wild Comb in Top Feeder Comb Spacing Comb Spacing 30mm

Brood nest that moved into a top feeder even with plenty of room in the boxes. Inner cover after removing the comb. Spacing on naturally drawn brood comb is sometimes as small as 30 mm but typically 32 mm.


How to get natural sized cells

Top Bar Hives..
Make the bars 32 mm (1 ¼") for the brood area
Make the bars 38 mm (1 ½") for the honey area

Foundationless frames.
Make a "comb guide" like Langstroth did (see "Langstroth's Hive and the Honey-Bee")
Also helpful to cut down end bars to 32 mm (1 ¼") or
Make blank starter strips
Use a brine soaked board and dip it in wax to make blank sheets. Cut these into ¾" wide strips and put in the frames.


How to get small cells

Use 4.9 mm foundation or
Use 4.9 mm starter strips


So what Are Natural Sized Cells?

I have measured a lot of natural drawn combs. I have seen worker brood in the range of 4.6 mm to 5.1 mm with most in the 4.7 to 4.8 ranges. I have not seen any large areas of 5.4 mm cells. So I would have to say:


Based on my measurements of natural worker brood comb:
o  There is nothing UNnatural about 4.9 mm worker cells.
o  5.4 mm worker cells are not the norm in a brood nest.
o  Small cell and natural cell have been adequate for me to have hives that are stable against Varroa mites with no treatments.

For an unbiased assessment of the health of my hives, here are health certificates from 2004 on.


Why mite counts don't matter
"All the boring and soul-destroying work of counting mites on sticky boards, killing brood with liquid nitrogen, watching bees groom each other, and measuring brood hormone levels—all done in thousands of replications—will someday be seen as a colossal waste of time when we finally learn to let the Varroa mites do these things for us...
"I have never yet counted even a single sample of mites from any of my bees. I consider counting mites as a way of evaluating Varroa resistance to be fraught with all sorts of shortcomings and difficulties. It's very time consuming and hence the size of the apiary, the number of colonies tested, the gene pool, and the income available all start to shrink. It's also very easy for the results to be skewed by mites migrating from other colonies or bee yards. "—Kirk Webster
"You don't grok the desert by counting the grains of sand."--Robert Heinlein, A Stranger in a Strange Land

If mites are reproducing on drones they are not damaging the colony much but they are more successful at reproducing.

Varroa count curve over the bee season on large cell worker comb with almost no drone comb will be low in the spring and high in the fall because the population increases somewhere between linear and exponential.

Varroa population curve over the bee season on natural comb with 20% drone comb will jump in the spring because the bees are raising a lot of drones and fall off whenever they are not rearing drones so it's lower in the fall.

If mites are reproducing on drones they are not damaging the colony as much.


Housel Positioning

I mention this because it is a topic to do with natural comb.

Primary CombOn BlankStarter Strip Typical PrimaryComb Confused Primary Comb Confused Primary Comb Closeup 4.9mm Foundation

Left, Housel positioning theory says they build a primary comb for the center. Here's a primary comb on a blank starter strip. This natural comb was brood and was drawn by regular Carniolans that were already regressed. The cells are 4.6 mm. The frames were spaced on 1 1/4" centers.

2nd from the left, my "Housel positioning" observations. This is a blowup of the same comb on the left. It is typical Primary Comb that I observe. Note the bar on the "Y" goes to the right. If you turn this around it's the same from the other direction. This is what I typically see for "primary" comb in natural comb hives. Housel positioning theory says the other combs should all be "Y" or inverted "Y" with the "Y" faces out and the inverted "Y" faces pointed toward the center primary comb. I have not been able to observe this pattern in my natural comb hives, but I have observed the typical primary comb. However I have had better acceptance and less burr following housel positioning with foundation.

3rd from the left, confused Primary Comb. Here is the first time I found a primary comb that wasn't the sideways "Y". Obviously no hard and fast rules on positioning.

4th, close-up of confused primary comb. Note the "Y" on the left and the upside down "Y" on the right and they are all on the same comb.

Far right a 4.9 mm deep foundation cut in half and put in a medium frame.


Frequently asked questions:


Doesn't it take longer for them to draw their own combs?


I have not found this to be true. In my observation (and others who have tried it), they seem to draw plastic with the most hesitation, wax with a little less hesitation and their own comb with the most enthusiasm. In my observation, and some others including Jay Smith, the queen also prefers to lay in it.


If natural/small cell size will control Varroa, why did all the feral bees die off?


The problem is that this question typically comes with several assumptions.

The first assumption is that the feral bees have all but died out. I have not found this to be true. I see a lot of feral bees and I see more every year.

The second assumption is that when some of the feral bees did die, that they all died from Varroa mites. A lot of things happened to the bees in this country including Tracheal mites, and viruses. I'm sure some of the survival from some of this is a matter of selection. The ones that couldn't withstand them died.

The third assumption is that huge numbers of mites hitchhiking in on robbers can't overwhelm a hive no matter how well they handle Varroa. Tons of crashing domestic hives were bound to take a toll. Even if you have a fairly small and stable local population of Varroa, a huge influx from outside will overwhelm a hive.

The fourth assumption is that a recently escaped swarm will build small cell. They will build something in between. For many years most of the feral bees were recent escapees. The population of feral bees was kept high by a lot of recent escapees and, in the past, those escapees often survived. It's only recently I've seen a shift in the population to be the dark bees rather than the Italians that look like they are recent. Large bees (bees from 5.4 mm foundation) build an in between sized comb, usually around 5.1 mm. So these recently swarmed domestic bees are not fully regressed and often die in the first year or two.

The fifth assumption is that small cell beekeepers don't believe there is also a genetic component to the survival of bees with Varroa. Obviously there are bees that are more or less hygienic and more or less able to deal with many pests and diseases. Whenever a new disease or pest comes along the ferals have to survive them without any help.

The sixth assumption is that the feral bees suddenly died. The bees have been diminishing for the last 50 years fairly steadily from pesticide misuse, loss of habitat and forage, and more recently from bee paranoia. People hear about AHB and kill any swarm they see.


If bees are naturally smaller why didn't anyone notice? Also why are the bee scientist saying they are larger?


I don't know why, perhaps some of it comes back to the regression issue. If you take bees from large cell comb and let them build what they want, what will they build? Is this the same as natural comb? Sometimes we just have differences in observations because of a variety of factors being involved.

I really don't think it should be hard to accept that they are naturally smaller since there have been plenty of measurements taken over the centuries. Dee Lusby's writings have references to many articles and discussions on the size of bees and comb and the concept of enlarging it. We have plenty of easy to find evidence that bees used to be smaller.

Find ABC & XYZ of Bee Culture books and look under "Cell Size".

Here's some quotes from them:

ABC & XYZ of Bee Culture 38th Edition Copyright 1980 page 134

"If the average beekeeper were asked how many cells, worker and drone comb, there were to the inch, he would undoubtedly answer five and four, respectively. Indeed some text books on bees carry that ratio. Approximately it is correct, enough for the bees, particularly the queen. The dimensions must be exact or there is a protest. In 1876 when A.I. Root, the original author of this book, built his first roll comb foundation mill, he had the die faces cut for five worker cells to the inch. While the bees built beautiful combs from this foundation, and the queen laid in the cells, yet, if given a chance they appeared to prefer their own natural comb not built from comb foundation. Suspecting the reason, Mr. Root then began measuring up many pieces of natural comb when he discovered that the initial cells, five to the inch, from his first machine were slightly too small. The result of his measurements of natural comb showed slightly over 19 worker cells to four inches linear measurement, or 4.83 cells to one inch."

Roughly this same information is in the 1974 version of ABC and XYZ of Bee Culture on page 136; the 1945 version on page 125; the 1877 version, on page 147 says:

"The best specimens of true worker-comb, generally contain 5 cells within the space of an inch, and therefore this measure has been adopted for the comb foundation."
All of the following historic references list that same measurement, 5 cells to the inch and can be reviewed at Cornell's Hive and the Honey Bee Collection online:
  • Beekeeping by Evertt Franklin Phillips pg 46
  • Rational Bee-keeping, Dzierzon pg 8 and again on pg 27
  • British Bee-keeper's Guide Book, T.W. Cowan pg 11
  • The Hive and the Honey Bee, L.L. Langstroth pg 74 of the 4th edition but is in all of them

This is followed in all but the 1877 version of ABC XYZ, by the way, with a section on "will larger cells develop a larger bee" and info on Baudoux's research.

So let's do the math:

Five cells to an inch, the standard size for foundation in the 1800s and the commonly accepted measurement from that era, is five cells to 25.4mm which is ten cells to 50.8 mm. This is 3.2 mm smaller than standard foundation is now.

A.I. Root's measurement of 4.83 cells to an inch is 5.25 mm which is 1.5 mm smaller than standard foundation. Of course if you measure comb much you'll find a lot of variance in cell size, which makes it very difficult to say exactly what size natural comb is. But I have measured (and photographed) 4.7 mm comb from commercial Carniolans and I have photographs of comb from bees Pennsylvania that are 4.4mm. Typically there is a lot of variance with the core of the brood nest the smallest and the edges the largest. You can find a lot of comb from 4.8 mm to 5.2 mm with most of the 4.8 mm in the center and the 4.9 mm, 5.0 mm and 5.1 mm moving out from there and the 5.2 mm at the very edges of the brood nest.

"Until the late 1800s honeybees in Britain and Ireland were raised in brood cells of circa 5.0 mm width. By the 1920s this had increased to circa 5.5 mm."-- John B. McMullan and Mark J.F. Brown

The influence of small-cell brood combs on the morphometry of honeybees (Apis mellifera)--John B. McMullan and Mark J.F. Brown

The 41st edition of ABC XYZ of Bee Culture on Page 160 (under Cell Size) says:

"The size of naturally constructed cells has been a subject of beekeeper and scientific curiosity since Swammerdam measured them in the 1600s. Numerous subsequent reports from around the world indicate that the diameter of naturally constructed cells ranges from 4.8 to 5.4mm. Cell diameter varies between geographic areas, but the overall range has not changed from the 1600s to the present time."

And further down:

"reported cell size for Africanized honey bees averages 4.5-5.1mm."

Marla Spivak and Eric Erickson in "Do measurements of worker cell size reliably distinguish Africanized from European honey bees (Apis mellifera L.)?" -- American Bee Journal v. April 1992, p. 252-255 says:

"...a continuous range of behaviors and cell size measurements was noted between colonies considered "strongly European" and "strongly Africanized". "
"Due to the high degree of variation within and among feral and managed populations of Africanized bees, it is emphasized that the most effective solution to the Africanized "problem", in areas where Africanized bees have established permanent populations, is to consistently select for the most gentle and productive colonies among the existing honey bee population"

Identification and relative success of Africanized and European honey bees in Costa Rica. Spivak, M

Do measurements of worker cell size reliably distinguish Africanized from European honey bees (Apis mellifera L.)?. Spivak, M; Erickson, E.H., Jr.

In my observation, there is also variation by how you space the frames, or variation on how THEY space the combs. 38 mm (1 ½") will result in larger cells than 35 mm (1 3/8") which will be larger than 32 mm (1 ¼"). In naturally spaced comb the bees will sometimes crowd the combs down to 30 mm in places with 32 mm more common in just brood comb and 35mm more common where there is drone on the comb.

So what is natural comb spacing? It is the same problem as saying what natural cell size is. It depends.

But in my observation, if you let them do what they want, for a couple of comb turnovers, you can find out what the range of these is and what the norm is. The norm was (and is) not the standard foundation size of 5.4 mm cells and it is not the standard comb spacing of 35 mm.

Studies on Small Cell:

I will try to update this from time to time as the links get broken. Please let me know if they are not working. Here are some studies you may find interesting:

Discussions on studies:

Michael Bush

email address

Copyright 2005-2006 by Michael Bush

Bush Bees Home