Although I was unable to confirm that robbing screens reduce the immigration of mites into a hive, I wondered about their various designs, and how they affected the foragers of the colony.

Introduction and Some Scientific Terms

Honey bee colonies exhibit a behavior termed “larder caching” –– the storing of food reserves. Their energy-rich larder of honey is highly attractive to predators and thieves, so honey bees seek sturdy domiciles with a defensible opening, have developed strong guarding and defensive behaviors, and back them up with weaponry (if needed).

The thievery of food from another animal is termed “cleptobiosis.” Honey bees routinely practice conspecific cleptobiosis (stealing from one’s own kind), and likely steal more honey from other colonies than do bears, ants, and other predators combined. Such cleptobiosis may be overt (robbing en masse (Figure 1)) or covert (the commonly-observed “surreptitious,” “progressive”, or “silent” robbing by foragers that “sneak” in and out of other hives).

Fig. 1 We’ve all seen overt robbing taking place. Note that these robbers are attracted to the scent issuing from a crack in a loose lid, rather than visually searching for an opening elsewhere.

Honey bees also practice “nest usurpation,” when a swarm simply moves into another colony’s domicile, kills its queen, and “takes over” ownership of the invaded colony’s hard-earned stores.

Practical application: Beekeepers routinely do this, when they requeen a colony, or combine two colonies.

Don’t Anthropormorphize Honey Bees

Making a living as a nectar forager is hard work. But Mother Nature has no rules, and all’s fair in the competitive game of evolution. Unlike humans, honey bees don’t need to justify their plundering of the less powerful or unfortunate, since all they are doing is a form of foraging. As noted by Breed [[1]]:

The western honey bee is a feeding generalist and forages on both floral and nonfloral sources of sugars, such as extrafloral nectaries. The evolutionary switch from foraging on nectar collected at flowers and extrafloral nectaries to foraging on stored honey in other bees’ colonies is within the foraging flexibility of honeybees and may not have required any particular adaptations to allow bees to make this switch.

My point is that we should avoid anthropomorphize bees; “robber” bees are not engaging in any moral transgression when they help themselves to unguarded sources of sugar –– they are merely foraging, and if they’re not chased away from a food source, they help themselves. This happens every day on flowers, when another bee of a different species may scare them off a blossom. If a colony does not actively defend its hard-won stores, their nectar or honey is fair game for the foragers of another colony, with no evil intent on their part.

Practical application: Robbery is defined as the action of taking property by force. Robber bees are thus misnamed –– they are actually only thieves, since it is only the guard bees that engage in fighting. Robbers may push and squeeze each other once overt robbing has started, but initially they act as thieves very much attempting to avoid a fight.

An Example of Robbing Without a Fight

I’m typing these words in early September, after checking on a tiny swarm that I’d hived a couple of days ago (swarms are normally unusual during our late-summer dearth, but for some unknown reason, I’m seeing a few swarms issue even from weak hives). The day after I hived it into a nuc, I gave it a half gallon of syrup in a top feeder. Several hours later, I noticed unusual activity at the entrance, but no fighting. I of course suspected that foragers from other hives were entering to help themselves to the syrup. Since no fighting was taking place, out of curiosity I allowed the apparent “gentle robbing” to continue.

The next day, the syrup bottle was empty, and flight at the entrance was back to normal. So I opened the nuc to see what had happened. The tiny cluster was still there, but there was barely a drop of syrup stored in the combs. It was pretty clear that every drop of that half gallon of syrup had been carried away by foreign bees –– without a fight, since there were no dead or injured bees to be seen.

I’ve observed this sort of “robbing” without fighting again and again, sometimes with nucs during a spring nectar flow, or when a colony weakened by mites or disease is unable (or chooses not to) defend itself –– it may be in the self-interest of outgunned workers to stand aside and allow the robbers to plunder their larder without a fight, since robbing bees have no interest in fighting, and disappear once they’ve cleaned out the combs, leaving behind the weak colony unharmed, other than having lost its stores.

There’s Little Difference Between Foraging and Robbing

The most common source of precious sugars for honey bees is the nectar of flowers. Foragers identify resource-rich flowers not only by sight (color and shape), but especially by their odors (flowers did not evolve their fragrances to woo humans). As any beekeeper with a honey house knows, the odor released when one uncaps combs of honey is extremely attractive to foragers –– especially once they’ve had a taste of it and identified that odor to be associated with a high-sugar reward. A lucky forager will immediately report its discovery to other foragers in the hive via the waggle or round dance, and soon we may see thousands of “recruits” seeking the specific odor of that sugar source everywhere in the vicinity.

Ditto with feeding bees syrup –– any spilled syrup, especially if scented, quickly attracts foragers, which if they detect that odor coming out of a hive ripening that syrup, tempts them to enter that hive.

Practical application: Having foragers merely investigating unguarded honey or syrup is not a problem for the beekeeper, unless one of them gets lucky and successfully returns to its hive with a rewarding crop full of sugar, and then recruits additional foragers to join it in harvesting the remainder (before another colony discovers it). Those recruited foragers go out searching for its associated odor (and perhaps its geographic location if far enough from the hive that the discoverer performed the figure-8 waggle dance). My point is that there’s no way for the discoverer to tell the recruits whether that reward came from a flower, a leaking syrup nozzle, a box of extracted combs, or an active guarded hive –– so the recruits (and additional scouts) are not visually looking for a specific object, but rather “sniffing” for a specific odor. Honey bees live in a world of smells.

How Robbing Screens or Guards are Presumed to Work

I was surprised that I was unable to demonstrate that robbing screens reduced mite immigration via the drifting of bees from high-mite hives to low-mite hives (this opposed to immigration on returning robbers, which would presumably not be affected by a screen).

The concept of how robbing screens work is founded on two assumptions: (1) that the fanned ventilation air from a hive will predominately exit straight out horizontally through the screen, rather than going up through the small offset entrance, and (2) since potential thieves are attracted to scent, they will presumably attempt to enter the hive through the aromatic screened area on the face, and “give up” when they are unable to find the offset entrance.

Practical application: Be careful with assumptions and conjecture! Although the above assumptions seemed plausible, it occurred to me that both could be tested for validity.

Test 1: Does any exhaust air exit out the upper entrance?

Bees actively ventilate their hive, especially if ripening nectar or cooling the hive in hot weather. It occurred to me that any exhaust air exiting a hive would normally be less dense than the ambient air, due to two reasons:

Warm air is less dense than cooler air. Colonies generate heat, especially when ripening nectar, and maintain a broodnest temperature of around 95°F (35°C), so the exhaust air will normally be warmer than the intake air, and tend to be more buoyant than the ambient air.
The exhaust air will generally also be of higher humidity than the intake air. Water molecules don’t evaporate into air –– they evaporate into space, and displace air molecules. Avogadro’s Law states that equal volumes of all gases (including air), under the same conditions of temperature and pressure, contain the same number of molecules. A molecule of nitrogen gas (N₂) has a mass of 28g, and O₂a mass of 32g, whereas a molecule of H₂O has a mass of 18g, so humid air will be considerably less dense than drier air, and likewise tend to rise.

So my question was, does enough of the less-dense exhaust air exit the top to guide a potential thief to that entrance? To find out, I set up an empty deep super with a bottom board and hive cover, and placed an aquarium pump inside to direct a weak stream of air towards the entrance. I then puffed slightly less-dense-than-air smoke from a smoker into the hive body and we watched where the exiting smoke went (Figures 2 – 5).

Our Findings

Fig. 2 Smoke exiting a box without a screen. Note how it rises, since it is slightly less dense than the ambient air –– simulating the odor-rich exhaust air normally exiting a hive.

Fig. 3 Smoke exiting a screen mesh guard. Note that the smoke (as proxy for the odor of ripening nectar) is also exiting through the top opening.

Fig. 4 Ditto for a metal guard with a very small offset upper entrance opening (black tape placed to better see the rising plume of smoke). Orientation to this odor plume would likely be how returning foragers eventually locate the new entrance –– so would potential thieves do the same?

Fig. 5 The same for a plastic guard. I prefer to call robbing screens robber guards, since the perforations in the metal and plastic ones hardly qualify as screens, since they would exhibit considerably more friction against the exhaust air than would a screen.

We found that although the aquarium pump put out a standardized exhaust stream, where the smoke exited the guards varied greatly with the air movement taking place outside the hive. But there was often a substantial amount of smoke (as a proxy for odor) exiting the upper entrances of the robbing guards. Since potential thieves investigate odor sources rather than visual locations, it was clear that an investigator (or returning resident bee) might well orient to the plume of exhaust air exiting the top entrance.

Practical application: To deter investigating nectar thieves orienting to an odor, it appears that the common designs of robbing screens could possibly stand improvement. To quote from themeshcompany.com: A woven wire mesh’ open area percentage measures how much void space there is within the structure, showing how easily air… can pass through. For designers…calculating this percentage isn’t just an academic exercise, it’s a necessity. Not only that, but adding a simple angled baffle inside the guard could likely minimize the escape of exhaust from the top entrance, without causing much additional restriction of bee passage. But as we’ll see in a moment, a guard’s efficiency at diverting potential robbers comes at a cost.

Test 2: If returning resident bees can eventually figure out how to get in, why wouldn’t potential robbers?

It also occurred to me that the success of a robbing screen is based upon the premise that a foreign potential intruder would not persevere as much as a returning resident bee at figuring out how to get in (perseverance being the key word).

When one installs a robbing screen on a hive, the colony’s returning foragers initially attempt to reenter the hive through the faceplate screen –– even if they had exited through the upper escape (Figure 6).

Fig. 6 Upon placement, even though bees exit via the new top entrance, they return to the bottom — presumably due to the outgoing colony odor.

But eventually, you’ll see foragers entering through the “hidden” entrance” (Figure 7).

Fig. 7 The colony’s foragers eventually learn where the new entrance is (some more quickly than others). However, I always see some bees at the screen face, and wondered why (which kicked off our subsequent investagtions below).

The assumption is that all the foragers eventually learn the new entrance, since by nightfall there are no more bees outside the entrance and we don’t see unsuccessful dead bees lying in front of the hive. But that’s only an unfounded assumption –– they could have given up and drifted to another hive (or flown off to die).

But if the foragers can all figure out the new entrance, why wouldn’t potential thieves? To see what actually occurs, we plucked 100 guard bees from the front of a hive at dusk and glued steel discs to their back, let them all return to the hive, and then placed a robbing screen on the face of the hive. I took a break to snap this photo of Rose intensely concentrating on applying a drop of glue (to avoid touching a wing) (Figure 8).

Fig. 8 We plucked 100 presumably guard bees from the hive entrance, glued steel tags on their backs, and let them walk or fly back into the hive. Watch a video of me tagging a bee at [[2]].

I wanted to minimize the variable of which bees we arbitrarily chose to tag, targeting bees that:

Hadn’t yet been foraging (and thus trained to visually orient to the “normal” hive entrance).
Would be taking their first foraging flights — having first exited through the new top-of-the-guard entrance.
Taking that first flight within the next few days after tagging.

It occurred to me that any bees facing us from the periphery of the cluster would fit the bill, since they would be exhibiting guarding behavior. Moore and Breed found that a worker bee normally engages in guarding behavior for only a day, and then shifts to foraging [[3]] (Figure 9).

Fig. 9 For all runs after the first, we plucked guard bees from the top bars after removing the hive cover, and performed the tagging a few inches away on top of the hive cover laid crossways . Ater gluing a tag on a bee, we released it to walk or fly the few inches back to the to the combs. To our surprise, instead of hiding after the indignity of being man-handled and tagged by us (as gently as possible), after diving back down between the combs, many reappeared right back up to the same place that we had originally plucked them from. And when we reached between the top bars to pluck a new guard be facing us to tag, they often came out with a tag already on them! By the time we had tagged a hundred, their behavior clearly confirmed that they were guards (although they rarely flew towards us or attempted to sting). In this photo, you can see 7 bees with tags taken at the time of tagging the last few bees (we were using up a mixed batch of previously-painted tags).

We placed magnetic traps [[4]] at the entrances to the hives on either side of the tagged-bee hive, each hive about a foot distant (Figure 10).

Fig. 10 Layout of our first run of the experiment. In order to quantify the effect of a robbing guard upon the drifting of bees taking their first foraging flight, we used three hives next to the house (already being used in a pollen sub experiment, and thus having viewing covers above). Rose is showing the placement of a plastic robbing screen; the hives to either side have magnetic traps above their entrances.

The day after tagging it was easy to observe tagged bees at the entrance, trying to figure out how to “get in” (watch a video at [[5]]). Many flew off in apparent frustration and soon returned to try again (often multiple times), but others didn’t return while we watched. We used the magnetic traps to determine if any frustrated tagged bees “gave up” and drifted into adjacent hives (the traps lift an entering tagged bee off its feet; it then struggles briefly to free itself –– leaving its tag on the magnet). We were stunned by the number of tagged bees that drifted to a hive next door –– 33 of the 100 did so in the first two days!

(Even more amazing is that a few days later, I realized that we had undercounted due to a screwup. We had not tagged bees for a few years, and in the confusion of locating the sheets of metals that we had initially experimented with, we had inadvertently punched an unknown proportion of the 100 discs out of aluminum! I only realized this a few days later when I observed a tagged bee walking out through the magnetic trap. So the 33% percent drift was certainly an undercount.)

Was this astounding amount of bee abandonment of their hive an anomaly, or was this the normal effect of a robbing guard? So we repeated the experiment again –– once with the same hives, and then three more times in another apiary containing 25 hives, placing magnetic traps on the 24 hives without the tagged bees and robber screen (each time with a different test hive, choosing ones located near the center of the apiary).

And finally, we reused the original hives at home to perform a Control run with no robbing screen, to determine the amount of “normal” drift to be expected after tagging 100 bees.

Our Findings

I’ve summarized the results in Table 1.

In all cases, tagged-bee drifting dropped off after two days. But we couldn’t assume that all the tagged bees had actually exited the hive. To check, four days after placing the guard on the 9 Sep run hive, we placed a magnetic traps over its entrance for three days to recover any additional flying tagged bees still in the hive (I then checked inside and found 3 that have never flown). We recovered 56, indicating that nearly 60% the tagged bees had been successfully flying in and out of the robber screen during the magnetic recovery period (in total, we recovered 83 of the 100 tagged bees on the magnetic traps, plus a few on the bottom board, with no idea of what happened to the others).

In the 22-Sep Control run, only two tags showed up (both in the first hour after tagging, which was unusual). No additional tags were recovered over the next two days, despite me encouraging foraging by putting out scented syrup, and “chumming” the colony. An hour later I observed a number of tagged bees mixed in the frenzy [[6]]. These observations support the conclusion that that there is far greater drifting of exiting tagged bees if a robbing screen is placed over the entrance.

In the outyard trials with lots of surrounding hives, nearly all the drifted tags were recovered from hives immediately or fairly close to the test hive, indicating that frustrated returnees are more likely to drift if there is a closely-adjacent hive. Of interest were the large number of returnees still trying to get in through the wire mesh of the 16-Sep run when I took drift counts three days after installing the screen (Figure 11), suggesting that this type of robbing screen may be more confusing to the bees than the plastic one –– which might make it more effective at deterring potential thieves.

Fig. 11 Was the reason that there was still so much confusion at the entrance of this mesh screen after three days due to there not being an adjacent hive on one side? Or was it because the mesh was better at confusing returning bees?

Based upon our observations of tagged bees attempting to reenter their screened hives, many attempt to find a way in, but then fly away in frustration, only to soon return to try again. But during that flyaway period, it appears that some may expand their search range and drift into a nearby hive entrance. But the majority are eventually successful in figuring out how to get back into their new hive entrance.

Practical application: Our findings indicate that when you place a robbing screen on a hive, that up to a third of the young bees in that hive may drift to other hives when they first exit! On the flip side, up to 85% may “figure out” how to find a way in –– suggesting that a determined robber might do so too!

Discussion

Robbing screens can clearly frustrate bees trying to enter a hive –– not only potential thieves, but also the workers of that hive. It was a complete surprise to us that placement of robbing screens may cause a substantial amount of workers to drift to other hives.

Practical application: The tradeoff for robbing screen design is that the better device thwarts robbers, the more it will likely also thwart first-time flyers from getting back into the hive.

There’s no reason to think that the roughly 25% figure for bees drifting after their first exit would not continue with every ensuing generation of workers taking their first flight, so long as the robbing screen remained on the hive. Think about it –– if there are 1000 workers reaching forager age each day (not unusual during robbing season), that could be 250 bees lost from the hive on their first flight –– every single day! Most would likely drift to a nearby hive, but if that hive also had a robbing screen on it, how many would starve before they found a way in? And how many would drift if there were no nearby hives were to receive them?

Practical application: Beekeepers often install robbing screens on weak hives. What impact would the loss of 25% of first-time foragers have upon that colony’s buildup and foraging each day? And what if its weakness were due to high varroa levels, or another transmissible pathogen –– would a robbing screen increase transmission to the other hives that receive the drifted bees?

Our findings suggest that more research needs to be done to determine whether the beneficial effects of robbing screens outweigh their possible negative effects.

Next Month

Our curiosity aroused, we next investigated robbing behavior and ran experiments to determine to what extent robbing screens discourage potential robbers (thieves) from entering an aromatically-attractive hive.

Acknowledgements

Thanks to Dr. Norm Gary for inventing the steel tag technique and advising me. And to my helpers Rose Pasetes and Corrine Jones.

Citations and Notes

[1] Breed, MD, et al (2012) Cleptobiosis in social insects. Psyche: A Journal of Entomology 2012(1): 484765.

[2] https://youtube.com/shorts/pyJv0BavFaA?feature=shared

[3] Moore, AJ, et al (1987). The guard honey bee: ontogeny and behavioural variability of workers performing a specialized task. Animal Behaviour 35(4): 1159-1167.

[4] Fig. 8 in https://scientificbeekeeping.com/8063-2/ and Fig. 2 in https://scientificbeekeeping.com/a-study-on-bee-drift-and-mite-immigration-part-4/

[5] https://youtu.be/UNydqwcJ7-I

[6] See at photo at https://scientificbeekeeping.com/scibeeimages/Feeding-frenzy-tags-537×450.png

Category: Varroa IPM Strategies, Varroa Management