Wild or Feral? A Debate About Free-Living Honey Bees

Why do news headlines so often use the word swarm to describe defensive bee behavior? It’s a disservice to the fascinating subject of reproductive swarming—and to the word swarm itself. Fortunately, when you read beyond the headline, you’ll usually find more balanced reporting, often including an interview with a local beekeeper.

Now, let’s turn to a debate that really gets academics buzzing: Should unmanaged honey bees living outside the species’ native range be called wild or feral?

In the November 2026 issue of American Bee Journal, Michael Thiele makes the case that established, self-sustaining honey bee populations living without human management should be called wild, not feral. Who knew a couple of words could stir up so much discussion?

Thiele writes:

“Feral” marks a member of a domesticated species living free of human control…. The domestication premise is itself the weak link. The heredity of the honey bee has remained under natural selection, the species carries no distinct breeds, and breeding effects dissipate through open mating.

That got me thinking about my own beekeeping. Perhaps bee landlording is a better description than domestication. I provide the housing, but the bees are free to leave and mate as they please. My backyard apiary isn’t surrounded by a screened barrier or a large body of water that would isolate the bees from other populations. How much control do I really have over their genetics?
So, what’s the difference between wild and feral?

Here’s how the terms are used in the article:

Term	Meaning
Wild	A self-sustaining population that reproduces over multiple years without depending on swarms from managed colonies. It lives independently of beekeeping interventions such as feeding and treatment, with traits shaped by natural selection—including, potentially, tolerance to Varroa destructor and its associated viruses.
Feral	Recently escaped colonies that have not yet demonstrated independent, long-term persistence and may still depend on new swarms from managed colonies to maintain their presence.

The distinction is more than semantics. It raises questions about how we define domestication, how natural selection shapes honey bee populations, and whether bees living beyond their native range can become genuinely independent of human care.

For those of us interested in free-living honey bees, these questions are especially intriguing. Are we looking at escaped descendants of managed colonies, or populations that have adapted to survive and reproduce on their own?

And perhaps my bee landlording arrangement is less about managing bees than providing them with an opportunity to make their own choices.
Want to explore the topic further?

If you’re interested in the history and evolution of beekeeping terminology, free-living bees, or the fascinating practice of bee-lining, I highly recommend these podcast interviews with Jamie Ellis and Kaylin Kleckner.

Explore the Arboreal Apiculture Salon recordings

You may find yourself looking at the bees in your own backyard a little differently!

Does UV-Blue Really Make a Better Swarm Trap?

I was first introduced to Brian Fleishmann’s September 2022 Bee Culture article, “Impact of UV-Induced Blue Fluorescence Entrances in Bee Culture,” through the MABA swarm trap build group. I was quickly convinced to give it a try. In 2023, I purchased four UV-reactive blue swarm trap entrances from the ApiTekStore. Not everyone in the swarm trap-building community was convinced. Brian’s experiment points toward a positive result, but with only six observations, the evidence isn’t strong enough to call the results statistically conclusive.  Still, I keep coming back to Brian’s experiment because, regardless of the statistical limitations, I think the experimental design is beautiful.

Brian’s article came to two conclusions:

• UV-reactive blue swarm trap entrances nearly doubled the number of scout bees observed compared with the control entrances.
• Swarms selected the trap with the UV-blue entrance over the control 100% of the time.

Those are pretty compelling results for a beekeeper looking for an edge in swarm trapping. Before getting too excited, it’s worth looking at how the experiment was conducted.

	Scout Bee Activity Logged	Trap Selected
Year 1 - 2020
Location 1	Yes	UV-reactive blue
Location 2		UV-reactive blue
Location 3		UV-reactive blue
Year 2 - 2021 (trap entrance swapped)
Location 1	Yes	UV-reactive blue
Location 2		none
Location 3		UV-reactive blue

Brian conducted the experiment from May through August in 2020 and 2021 in western New York.
The basic setup:

• Traps were attached to trees 8–10 feet above the ground.
• Each trap had a 1.25-inch (3.2 cm) circular entrance hole.
• Trap cavity volume was 1.59 cubic feet (45 liters).
• Each trap contained four frames without comb and one frame with some comb.
• Three locations faced open fields with bee forage.
• Two traps were placed at each location, approximately 200 feet apart.
• One trap had a 3D-printed entrance made from UV-reactive blue filament.
• The other used white filament as the control.
• Swarm lure spray was applied weekly.
• Weather permitting, one location was monitored for scout bees between 11:30 a.m. and 4:00 p.m.

There was also a particularly clever element to the 2021 experiment: the UV-reactive and control entrances were swapped between traps.  Why? To control for the possibility that the bees simply preferred one particular tree. That detail is one of the reasons I like the experiment so much. Brian wasn't just asking, “Do bees like blue?” He was trying to eliminate other explanations for the bees' behavior. 

The biggest weakness is the sample size.
With only six observations, the results don't provide the kind of statistical confidence needed to make a definitive claim that UV-blue entrances attract more swarms. A larger experiment with more traps, locations, and swarm events would be much more convincing.  

To see whether Brian's findings had any support from other research, I asked Perplexity to look for academic, peer-reviewed citations of his article. Unfortunately, it couldn't find any. That isn't necessarily surprising. Brian's article appeared in Bee Culture, a trade publication, rather than an academic journal.
Perplexity did uncover a related 2018 study by Ostroverkhova et al., “Understanding innate preferences of wild bee species: responses to wavelength-dependent selective excitation of blue and green photoreceptor types.” Brian also references this study in his article.  Ostroverkhova found that significantly more bees were captured in their “attractive” traps equipped with blue fluorescent vanes than in the other traps they tested, with a statistically significant result of P < 0.05.  That doesn't prove Brian's swarm-trap experiment was correct. The experiments were not identical, and we shouldn't treat them as interchangeable.  It does provide an interesting piece of supporting evidence: bees can show a measurable attraction to particular blue wavelengths.  So while Brian's experiment has limited statistical power because of its small sample size, its findings don't exist in complete isolation. They point in a direction that is also supported by a separate experiment with stronger statistical evidence.
 

Fast-Forward to 2026
My original UV-reactive PLA swarm trap entrances from ApiTekStore have now been through several seasons, and they're showing their age. The plastic has become brittle and the blue color has faded.  Then, in April 2026, I met MABA speaker Ryun Forsman, better known as The3DBee, who gave a presentation on 3D printing.  Ryun makes swarm trap entrances using durable PETG filament, and his design has a much larger footprint along with a stronger UV-reactive response than my original entrances.  I bought six and I’m putting them to work.

Here's a short video of scout bees visiting one of my traps on March 10, 2026.

On April 13 2026, at 12:30 PM all the scouts vanished and shortly after the swarm started to move in.  Here’s a time lapse video of the swarm arrival. 

There Is No Perfect Swarm Trap Recipe
At this point, I'm comfortable saying that UV-blue entrances add value to my swarm trap management.  There are many best-practice swarm trap variables that may never have a settled answer.  For example, I'll share two things I do differently from Brian's experiment:

1. I use a slow-release Nasanov pheromone vial.
2. I don't put old comb in my swarm traps.

Despite those differences, neither practice appears to have hurt my year-over-year success.  Swarm trapping isn't about finding a single magic recipe. It's about stacking the odds in your favor.

a web of interconnected ideas

I had originally imagined writing a blog post centered on numbers, but after hearing MABA speaker Nathalie B., reflecting on the reactions of beekeeping students at the Atlanta Transition Center, and reading Paul Honigmann’s book, I found myself heading in a different direction. Instead, I want to spend a little time thinking about how confident we are—or think we are—as beekeepers.

The Dunning–Kruger Effect   As we progress from classroom instruction to hands-on beekeeping and eventually to managing colonies on our own, our self-assessment of our confidence changes in ways that don't always match our competence.

On the left side of the graph is the overconfident novice. Imagine someone who has just learned about culling queen cells and suddenly sees queen cells as the solution to every problem—a hammer in search of nails.

The valley of despair might come when that same enthusiastic beginner excitedly shares their new culling knowledge with a more experienced beekeeper, only to be met with surprise or skepticism. It's an uncomfortable moment, but often an important step in learning.

A little farther to the right is the beekeeper who has gained some experience but is still developing competence. They arrive at the apiary with a clear plan for manipulating a hive, only to discover that the bees have different ideas. Instead of adapting, they struggle because the hive doesn't match the script they had written in their head.

On the far right is the seasoned beekeeper. This is someone who reads widely, belongs to their local beekeeping club, and has developed preferred management practices through years of experience. At the same time, they are comfortable listening to viewpoints that differ from their own, knowing there is rarely a single correct answer. For this end of the graph, I've chosen the label "It depends." During her treatment-free presentation at MABA, Nathalie B. returned to that phrase again and again. It depends on the genetics. It depends on the forage. It depends on whether the bees are locally adapted. It depends, it depends, it depends.

Perhaps that's the clearest sign of growing expertise in beekeeping: the realization that simple answers are rare, context matters, and confidence is best paired with curiosity.

continuous low-level mite erosion

A London beekeeper friend gave me two monthly magazines that members of the Bromley Beekeepers receive as part of their membership. One of them was BeeCraft, and while flipping through it, I came across an article by Paul Honigmann.  The name caught my attention. Honigmann literally translates to "honey man," which seems wonderfully appropriate once you start reading his work.

In his article, Paul highlighted a fascinating graph originally developed by Gareth John. It compared two very different approaches to managing varroa mites: a single, one-time treatment versus a continuous daily reduction of just 2% in the mite population. The graph was simple, elegant, and memorable. As a treatment-free hobby beekeeper, it immediately grabbed my attention. There was just one problem—for me, at least. The graph didn’t include the underlying numbers. And if you’re anything like me, that’s unsettling. I don’t just want to see the curve; I want to understand what drives it. Without the math, it felt incomplete… almost like an itch I couldn’t scratch.

So, naturally, I went digging.  With a little help from a large language model (Perplexity), I tracked down a practical growth model used by Randy Oliver. That gave me the missing piece—and from there, the rest fell into place.

Building the Model Behind the Graph

To recreate Gareth John’s insight, we can start with a simple assumption: in the absence of intervention, varroa mites grow exponentially.

M(t)=M₀e^rt

Where:

M₀ is the initial mite population

r is the daily growth rate

t is time in days

Randy Oliver suggests a reasonable rule-of-thumb growth rate of r=0.021 per day during active brood rearing. That corresponds to a doubling time of about 33 days.

For example, if a colony starts with 10 mites:

M(33)=10⋅e^0.021⋅33≈20

What Happens With Continuous Control?

What if the colony continuously removes a small fraction of mites—say 2% per day?

We can model that by adjusting the growth rate:

M(t)=M₀e^(r−0.02)t

Using the same r=0.021

M(t)=M₀e^0.001t

That’s a dramatically slower growth rate.   In fact, when you run the numbers, the population barely increases over time. It’s not quite flat—but it’s close enough to feel like balance.

A Different Way to Think About Control

This is where Gareth John’s insight really lands.  A steady, modest continuous reduction—just 2% per day—can almost stabilize the mite population. Compare that to a one-time treatment, which often leads to a rebound as mites continue their exponential climb.

It raises an interesting question: are the bees themselves already contributing to this continuous “background” mite suppression?  Gareth suggests they might be. And it’s easy to imagine other colony behaviors or environmental factors playing a role as well—hygienic behavior, grooming, brood interruption, or even subtle ecological pressures within the hive.

Kent Beekeepers

In densely populated London, it is not unusual to find beekeepers tending hives on rented sections of community garden plots, known locally as allotments. During a visit this May, I had the opportunity to explore the main apiary of the Bromley Beekeepers, located in the Sydenham area of southeast London.
Having visited many allotments over the years, I was surprised by the scale and energy of this site. The annual allotment open day felt more like a community festival than a gardening event. Visitors enjoyed craft and beverage vendors, live music, food sales (tea and cake) from the clubhouse, and sunny spring weather alongside hundreds of gardening enthusiasts and local residents.
One of the highlights of my visit was meeting several members of the bee club, including Lottie Simpson, creator of a remarkable mud-covered skep. The skep features a carved wooden mask inspired by bee anatomy, and a recently installed swarm has adopted the sculpture as its home, using the mask’s eye openings as entrances and exits. 

The Bromley Beekeepers organization offers members a range of benefits, including two national monthly magazine subscriptions and liability insurance. The visit offered a fascinating glimpse into London's thriving beekeeping community and highlighted the important role that allotments play—not only as spaces for gardening, but also as hubs for education, conservation, and community engagement.

My broader impression from observing pollinators in southern England is that the region appears to support a greater diversity of bumblebee species than I commonly encounter around Atlanta, Georgia. At first glance, this may seem surprising, since the southeastern United States is generally known for its rich biodiversity. However, bumblebees are particularly well adapted to temperate and cooler climates, making the United Kingdom an ideal environment for many species. While Georgia boasts an impressive diversity of bees, much of that richness comes from groups other than bumblebees, including solitary bees, sweat bees, leafcutter bees, carpenter bees, and numerous other native species. In fact, carpenter bees can easily be mistaken for bumblebees at first glance, as the two share a similar size and appearance, despite belonging to different groups.
Carpenter bees can look very similar to bumblebees at first glance.  Key differences 

Feature	Bumblebees	Carpenter bees
Social behavior	Usually social, living in colonies with a queen and workers	Mostly solitary (each female makes her own nest)
Nesting	Underground, in grass tussocks, compost heaps, cavities, etc.	Bore tunnels into wood, bamboo, dead stems, or similar materials
Abdomen appearance	Usually hairy/fuzzy all over	Often has a shiny, hairless black abdomen (especially large carpenter bees)
Colony size	Can range from dozens to hundreds of bees	Usually a single female or a small family group
Genus	Bombus	Xylocopa (large carpenter bees)
		Ceratina (small carpenter bees)

3 feet or 3 miles -- oversimplification

One of the most repeated maxims in beekeeping is the advice on moving hives: “Move them either 3 feet or 3 miles.” It’s simple, memorable… and, in my experience, not quite the full story.

This spring, two of my four hanging swarm traps successfully attracted swarms.

Swarm #1: The Neighbor’s Backyard Adventure
On March 22, a swarm chose a trap hung on a tree limb in a neighbor’s backyard. After letting them settle in for a couple of days, I lowered the trap to a more stable, level position. These traps held five deep, foundation-less frames, so keeping things level was important. My backyard hive stand is only about 540 feet away—far too close for a direct move without causing confusion among returning foragers—ask me how I know.

Using the MABA Bee Neighbor List, I found a fellow club member just over a mile away who was willing to host my temporary relocation. After dusk—bees safely inside—the trap rode shotgun (literally, balanced on my knees while my spouse drove) to its temporary home.

Five days later, I brought the trap back to my backyard hive stand, where it stayed another four days before I finally transferred the frames into a proper nuc box. The bees handled it beautifully.

Swarm #2: A Backyard Puzzle

The second swarm, on April 13, selected my backyard hanging swarm trap—just 50 feet from where I ultimately wanted them.

This time, instead of a long-distance detour, I moved the trap in 10 foot steps. Each move was planned carefully: I leveled the next location the night before, then moved the trap at dawn.

In total, it took five intermediate moves to reach the final hive stand.

What I observed was fascinating: a 10-foot move caused only minor confusion among the bees, and they seemed to reorient themselves within about four hours. No chaos, no mass disorientation—just a brief adjustment period.

Rethinking “3 Feet or 3 Miles”

After these experiences, I’m not convinced the classic maxim tells the whole story. It’s a helpful guideline, sure—but real-world conditions offer more flexibility than it suggests.

If I had to rewrite it based on what I’ve seen, it might sound more like:

“Move them less than 10 feet… or at least a mile away.”

Not quite as catchy—but perhaps a bit closer to reality.

tulip poplar windfall and phenology '26

Spring seems to be arriving a bit early this year—at least in my Atlanta backyard. My records show the first tulip poplar flower windfall appeared on day 84, about 13 days earlier than expected. It’s a small but telling sign that the season is shifting ahead of schedule.

To put this in perspective, I compared my observations with broader indicators like the National Phenology Network’s Spring Bloom Index. This model, based on flowering plants like lilacs and honeysuckles and seasonal temperature patterns, shows that the 2026 spring in Georgia’s Piedmont region is running a few days early compared to the 30-year average.

In other words, what I’m seeing locally matches the bigger picture—spring is getting a head start.

Other Atlanta nectar sources at this time (not shown):

• Chinese Wysteria (invasive)
  
• White Clover
• Spiderwort
• Amur Honeysuckle (invasive)
• Dandelion 

 

1st Tulip Poplar Windfall	Day of Year
10-Apr-2011	100
27-Mar-2012	87
19-Apr-2013	109
12-Apr-2015	102
15-Apr-2017	105
5-Apr-2018	95
11-Apr-2019	101
7-Apr-2021	97
6-Apr-2022	96
24-Mar-2023	83
6-Apr-2024	97
3-Apr-2025	93
25-Mar-2026	84
median	97