I'd like to thank these sponsors for supporting this website. Just click on their ads to go to their websites.

Mann Lake Ltd
Print Friendly

Sick Bees – Part 18F2: Colony Collapse Revisited – Keeping A Leaky Boat Afloat

First published in: American Bee Journal March 2013

Keeping A Leaky Boat Afloat

Keeping a Leaky Boat Afloat. 1

the leaky boat analogy. 1

Why would bees collect toxic pollen?. 2

Almond pollen. 4

colony-to-colony variation. 5

bee genetics or gut endosymbionts?. 5

summary (so far). 6



Randy Oliver

The Leaky Boat Analogy

One way to put pesticide impacts upon bee health into perspective would be to think of a bee colony as a wooden boat, with exposure to each individual toxin representing a leak, or in the case of insecticide dust or sprays, a wave over the bow.  The colony “boat” must work at “bailing out” the allelochemicals in nectar and pollen (by detoxification or elimination); the more phytotoxins coming in, the less safe freeboard remaining.  In some cases a phytotoxin load may overwhelm the bailing crew and sink the boat. 

In the above analogy, you can see that exposure to manmade pesticides is not a black or white situation—we must always keep in mind how much freeboard the boat had left prior to being exposed to the additional toxin.

Practical application:  so when we think of bee colonies and toxins of any sort, we must always keep in mind the amount of “bailing” that the colony is already doing to keep up with the leaks prior to its exposure to any new toxin.  And anything that adds to colony stress will likely make that colony more susceptible to damage by the addition of yet another toxicant.

On the other hand, the exposure of bees to some allelochemicals may prepare the crew for “bad weather”—by consuming plant toxins, the bees may upregulate their detoxification enzymes, kind of like the crew getting stronger by “practicing.”  May Berenbaum found that bees that consumed propolis extracts exhibited an enhanced capacity to detoxify mycotoxins [1].

Why Would Bees Collect Toxic Pollen?

There are two foods that foragers bring back to the colony—nectar and pollen.  However, the evaluation of the “safety” of these two food sources is entirely different.  In the case of nectar, there is a multi-step “homeland security” type of screening process to minimize the intake of toxins into the hive.  First, the individual forager samples the nectar, and may reject it should it find the taste repellent, since many phytotoxins are bitter or irritating.  Then on the flight home, should the forager become sickened by its nectar load, it may die in the field, or intentionally sacrifice itself by committing “altruistic self removal.”  Once back in the hive, the forager must then find a mid-aged receiver bee to take its nectar load, which provides yet another checkpoint against toxic nectar, since the receivers would likely be averse to taking any nectar that had previously bothered them.   So the colony appears have effective mechanisms in place to protect it from poisonous nectar.

Surprisingly, no such sort of screening appears to happen in the case of pollen!  Bees are well known to collect pollens toxic enough to kill the colony (as with California Buckeye in my area).  Since pollen foragers do not ingest the pollen that they gather, they have no way of knowing whether it is toxic or not.  Excellent studies by Pernal and Currie [2] demonstrated that foragers do not appear to discriminate between pollens due to nutritional content, but rather by particle size and aroma.  My own experimentation supports the importance of odor.

Since it would be non adaptive for bees to collect toxic pollen, one would suspect that natural populations of honey bees would evolutionarily  “learn” to avoid any toxic pollens in the local habitat based upon their odor, or to inherit the genes that caused them to favor pollens that were on the “proper odor list.”

So what happens when we move bees outside of their natural range, such as to the Americas, Australia, or New Zealand?  There they would be exposed to completely novel pollens, such as those from corn (maize), soybeans, melons and squash, sunflowers, cotton, blueberries, and cranberries (all of New World origin).  The European honey bee may not be well adapted to either the nutritional composition of these pollens, nor their allelochemicals.

  Not surprisingly, when researchers inspect the actual pollen loads of foragers from hives placed adjacent to these New World crops, they often find that the bees eschew those pollens in favor of those from plants of Old World origin growing within flight range (either weeds or crop plants).  I suspect that the bees may simply not be “wired” to recognize the New World pollens as food.  Perhaps not surprisingly, some of them are also of notably poor quality as bee feed.

Consider corn pollen.   In a number of studies, corn (maize) is often found to be one of the major pollen sources used by honey bees [3, 4], and in the U.S., bees eagerly gather the pollen of sweet corn.  Yet U.S. beekeepers have told me that bees winter poorly on corn pollen. 

These anecdotal reports are supported by recent research by Höcherl [5], who found that bees experimentally fed maize pollen produced less brood and had shorter longevity as adults than those fed mixed pollens.  Maize pollen is generally high in protein, but when I compared its amino acid composition to the ideal of deGroot [6], its nutritive value was limited by its low ratio of tryptophan.  However, that does not explain why bees so eschew the pollen of field corn.  Bromenshenk [7], who recently surveyed 116 colonies across the Corn Belt during the tasseling period, found that foragers preferentially seek out any other sort of pollen than that of field corn–a substantial proportion of colonies collected no corn pollen, and overall, the median percentage of corn in trapped pollen was only about 10%.  This finding means that foragers will fly over fields of field corn in full tassel in order to find weeds in flower.  So what’s up with that?

What could possibly make bees so dislike field corn pollen that the foragers would avoid gathering such an easily available and abundant food source?  Hungry foragers will gather sawdust (Fig. 1), so it can’t just be the lack of nutritional value.    


Figure 1. Bees collecting sawdust at my home yard on a sunny morning in January, when there was little natural pollen to be had. There was far more activity at a feeder of dry pollen supplement that I had set out.

 Could it be that allelochemicals in field corn pollen is repellent to foragers?  When I searched, I was surprised to find that the pollen of some varieties of corn “burns” the leaves of plants that it falls upon [8], perhaps due to its high content of phenylacetic acid (PAA).  I couldn’t find toxicity data of PAA [9] to bees, but it makes me wonder whether some of our cultivars of field corn are somewhat toxic to bees?

Almond pollen

One demonstrably toxic pollen of great interest to me is that of the almond tree (Fig. 2), which contains high concentrations of amygdalin (normally metabolically degraded to benzaldehyde and cyanide—also both toxic).  Yet bees thrive on almond pollen, despite the fact that in some cultivars the concentration of amygdalin approaches that of acute bee toxicity [10].  What concerns me is that no one has studied whether the necessary detoxification of amygdalin hampers the bees’ ability to simultaneously detoxify other pesticides.

Figure 2.  Almond pollen and nectar is loaded with the toxic and bitter cyanogenic glycoside amygdalin, yet honey bees thrive on it.  The bee is apparently preadapted to detoxify this class of chemicals since they are commonly produced by a number of European fruit trees in the rose family.

Figure 2. Almond pollen and nectar is loaded with the toxic and bitter cyanogenic glycoside amygdalin, yet honey bees thrive on it. The bee is apparently preadapted to detoxify this class of chemicals since they are commonly produced by a number of European fruit trees in the rose family.

Despite its apparent toxicity, bees thrive on almond pollen, presumably because of their coevolutionary adaptation in the Mediterranean region to the even more toxic ancestor of the cultivated almond.

Colony-to-Colony Variation

One can easily observe the colony-to-colony difference in response to the toxic organic acids and essential oils used for varroa control.  The exact same dose applied to a number of hives will elicit vastly different responses.  Some colonies appear to barely notice the chemical, whereas others may pour out of the entrance, suffer serious mortality, or even abandon the hive.  This fact makes it difficult to define the “toxic dose”!

We must also keep in mind that not every colony in an apiary is foraging on the same plants.  All one need do is to put pollen traps on each hive to observe that each colony can be foraging on completely different flowers, and perhaps in totally separate areas.  The exposure of two side-by-side colonies to toxins, whether natural or synthetic, may be 100% different! 

We tend to breed generic, highly productive bee stocks.  In natural settings, though, honey bees were much more locally adapted, fine-tuned to the toxins of the local flora.  On that subject, I found a fascinating and thought-provoking paper by Després [11], from which I’ll share some excerpts:

Over 400 million years of coevolution with plants, phytophagous [plant-eating] insects have developed diverse resistance mechanisms to cope with plant chemical defences.  Because insects face a geographical mosaic of chemical environments, from non-toxic to highly toxic plants, the costs associated with resistance traits vary with the probability of encountering a toxin…

I’ve heard anecdotally that in certain areas, locally-adapted bee stocks can better survive exposure to the local toxic flora (such as Buckeye in the Sierra foothills) or the dark fall honey that can cause winter losses in nonadapted stock.  Atkins [12] also pointed out that “Some hybrid strains of bees are more resistant to certain plant poisons than purer strains.”  This subject certainly deserves further investigation!

Bee Genetics or Gut Endosymbionts?

But is it actually the genetics of the bees that makes the difference, or the genetics of their symbiotic fungi and bacteria?  Honey bees harbor a distinctive microflora in both their guts and in the beebread [13].  In the beebread, the symbiotic fungi may be the more important for producing detoxification enzymes [14].  This hypothesis is supported by the finding by vanEngelsdorp [15] that pollen containing high levels of a fungicide apparently goes “toxic,” whereupon the bees seal it off.

I mentioned in my last article that pesticide detoxification appears to take place largely in the Malpighian tubules and ileum of the gut.  A recent study by Martinson [16] found that these organs appear to be specifically constructed to provide a home for symbiotic bacteria—perhaps those critical for the detoxification process.  These bee endosymbionts may be more important than we’ve realized:

Insect fungal symbionts (as well as bacterial symbionts) are well recognized as providers of nutrients for their insect hosts. The rationale behind this is that the insects are feeding on nutrient-poor resources, and need the microbial symbionts to provide a balanced diet. The effects of the presence of toxins in these food materials has been relatively overlooked, yet in many cases there is just as much need for detoxification as there is for nutrient provision [17] (emphasis mine).

It could be that a colony’s ability to handle pesticides is dependent upon the health and species composition of this gut community.  And what do we beekeepers often do?  We treat our colonies with antibiotics to kill bacteria!  Now I’m not about to claim that antibiotics are “bad” for bees—many beekeepers report that Terramycin (oxytetracycline) works like a “tonic,” similar to the manner in which poultry seem to benefit from antibiotics.  Indeed, the Moran lab [18] found that terramycin-resistant bacteria were common in the guts of commercial bees, but rare in colonies in which there was no history of antibiotic use.

But what happens if we introduce a novel antibiotic?  The authors noted that tylosin was registered for use by beekeepers starting in 2005 (and is widely applied applied in excess), just as CCD reared its ugly head.  Vásquez [19] found that the beneficial gut flora that they tested were more susceptible to tylosin than to tetracycline.  I had checked with Dr. Jerry Bromenshenk at the time of his first survey of potentials factors in CCD, asking whether there was any correlation with the use of tylosin—but none stood out.  

But there is more to the story.  Nectar contains not only plant allelochemicals, but also bacteria and fungi which can grow in the sugar-rich solutions and produce toxic metabolites.  Once bees consume nectar, their symbiotic gut bacteria suppress the growth of these potentially harmful nectar microorganisms.  Vásquez [20] collected some 55 bacteria and 5 fungi from flowers, and found that the most common bee endosymbiotic bacterium was also the most potent in inhibiting the growth of all the flower microorganisms.  Her findings suggest that the typical feeding of Tylosin in fall by commercial beekeepers may suppress these important gut endosymbionts, with unknown consequences. 

Practical application:  I’m not going to try to make a case against antibiotics–I use them myself when called for.  However, I will suggest that it would be wise to save them for when we really need them, rather than using them prophylactically across the operation.

Summary (so far)

Confused?  Me too!  We are still very much on the learning curve about bee responses to toxins!  Let’s review some points to keep in mind:

  • When we use the word “pesticides” or “toxins” we should realize that those terms include the plethora of natural plant allelochemicals to which bees are exposed via nectar, pollen, and propolis, and toxic bacterial and fungal metabolites in nectar and pollen.
  • At extremely low doses (parts per trillion), most plant allelochemicals and synthetic toxicants are likely merely “background noise” of little or no biological relevance. 
  • At very low doses many toxins clearly have beneficial hormetic effects, perhaps by initiating certain immune response cascades; or stimulatory effects (as do the natural alkaloids caffeine and nicotine).
  • I’ll spare you the math [21], but when I calculated out the amount of plant alkaloids that would be consumed by an individual bee in pollen or nectar, and compared it to the oral lethal dose as determined by Detzel, bees appear to be commonly exposed to what should be lethal levels of plant alkaloids.
  • It is typically only under certain environmental conditions that colonies show signs of poisoning from natural flora (I don’t normally notice Buckeye poisoning to any great extent).  Oh duh! [22] However, any sort of stress on those plants, or lack of dilution by “safe” nectar or pollen, could suddenly shift the balance and allow for the natural “poisoning” of colonies.
  • We have virtually no idea as to how important bee breeding and colony endosymbionts are in relationship to the bees’ ability to detoxify allelochemicals or synthetic toxicants.

To return to my leaky boat analogy, colonies foraging on toxic pollen or nectar may not have much freeboard left.  Add to that environmental pollutants, poor nutrition, or parasite load and such colonies will be unable to tolerate any additional manmade toxicants.  On the other hand, a colony enjoying good nutrition and maintaining a robust broodnest may be able to recover from a pesticide spray.

Practical applications: 

  • Learn if you have known toxic flora in your vicinity
  • Keep locally-adapted bees
  • Don’t count on corn, sunflowers, vine crops, blueberries or cranberries for nutrition
  • Mitigate the effects of toxins by supplemental feeding to promote broodrearing
  • Use antibiotics with caution


Next I’ll add the contribution of manmade pollutants and pesticides to the picture.


As always, I am indebted to Peter Borst, and to my wife Stephanie for her forgiveness for my long hours, and her help with the final edit.


[2] Pernal SF, Currie RW (2002) Discrimination and preferences for pollen-based factors by foraging honey bees (Apis mellifera L.). Anim Behav 63(2): 369-390.

Pernal SF, Currie RW (2001) The influence of pollen quality on foraging behavior in honeybees (Apis mellifera L.).  Behavioral Ecology and Sociobiology 51(1): 53-68.

[3] Keller I, et al (2005) Pollen nutrition and colony development in honey bees – Part II. Bee World 86: 27–34.

[4] Malerbo-Souza, DT (2011) The corn pollen as a food source for honeybees. Maringá, 33(4): 701-704. http://www.scielo.br/pdf/asagr/v33n4/20.pdf

[5] Nicole Höcherl, N, et al (2012) Evaluation of the nutritive value of maize for honey bees.  Journal of Insect Physiology 58: 278–285.

[6] deGroot, AP (1953). Protein and amino acid requirements of the honeybee Apis mellifera. Physiologia Comparata et d’Ecogia. 3: 195-285.

[7] Bromenshenk, JJ and C Henderson (2012) In press.

[8] Anaya, AL, et al (1992) Phenylacetic acid as a phytotoxic compound of corn pollen.  Journal of Chemical Ecology 18(6): 897-905.

[10] London-Shafir, I, et al (2003) Amygdalin in almond nectar and pollen – facts and possible roles. Plant Syst. Evol. 238: 87–95.

[11] Després (2007) Op. cit.

[12] Atkins, EL (1975) Injury to honey bees by poisoning.  In The Hive and the Honey Bee, Dadant.

[13] Martinson, VG, et al (2011) A simple and distinctive microbiota associated with honey bees and bumble bees. Molecular Ecology 20: 619–628. http://www.danforthlab.entomology.cornell.edu/files/all/martinson_etal_2011.pdf

[14] Gibson, CM and MS Hunter (2010) Extraordinarily widespread and fantastically complex: comparative biology of endosymbiotic bacterial and fungal mutualists of insects.  Ecology Letters 13: 223–234. http://ag.arizona.edu/ento/faculty/hunter/PAPERS/Gibson_Hunter_10.pdf

[15] vanEngelsdorp, D, et al (2009) ‘‘Entombed Pollen”: A new condition in honey bee colonies associated with increased risk of colony mortality. J of Invert Pathology 101: 147–149. http://ento.psu.edu/pollinators/publications/Entombed

[16] Martinson, VG, et al (2012) Establishment of characteristic gut bacteria during development of the honeybee worker. Appl Environ Microbiol. 78(8): 2830–2840. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3318792/

[17] Dowd, PF (1992) Insect fungal symbionts: a promising source of detoxifying enzymes. Journal of Industrial Microbiology. 9: 149-161.  http://naldc.nal.usda.gov/download/24916/PDF

[18] Tian, B, et al (2012) Long-term exposure to antibiotics has caused accumulation of resistance determinants in the gut microbiota of honeybees.  mBio 3(6): http://mbio.asm.org/content/3/6/e00377-12.full.pdf+html

[19] Vásquez A, Forsgren E, Fries I, Paxton RJ, Flaberg E, et al. (2012) Symbionts as major modulators of insect health: lactic acid bacteria and honeybees. PLoS ONE 7(3): http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0033188

[20] Vásquez, A, et al. (2012) Symbionts as major modulators of insect health: lactic acid bacteria and honeybees. PLoS ONE 7(3): http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0033188

[21] Typical food consumption by a nurse bee is about 10mg  of pollen per day.  Detzel quantified the alkaloids in lupine pollen at  about 50µg/g, so a bee consuming 10mg of pollen would be exposed to 0.5µg of alkaloids.  The LD50 for typical alkaloids (in syrup) was about 1/10th that amount.

[22]  As I’m reading these words to Stephanie for review, it occurs to me that this likely happened to my own colonies this year when the blackberry bloom failed to materialize!  We didn’t notice severe Buckeye poisoning, but the colonies sure didn’t do well.  It’s amazing to us that I can write an article, but have to read it aloud to realize that it may apply to me!

Category: Pesticide Issues