BRENNEN

Industries / Food, farming and forestry / Agriculture and farming / Beekeeping

Queen rearing and breeding

Raising new queens, getting them mated and into colonies, and choosing which colonies become the parents of the next generation.

7

Milestones achieved, from 1889 to 2006.

1

To watch for: announced, not yet confirmed outside.

2

Open deliverables. Nobody has shipped them.

The road

AchievedWatch forOpen

Achieved

  1. 1889

    Queens from wax cups

    A colony raises only a few queens on its own. In Scientific Queen-Rearing (Chicago, 1889), G. M. Doolittle described building wax cups by dipping a wooden stick into melted wax, putting royal jelly in each cup and moving young worker larvae into them, a transfer now called grafting, so that one colony could raise many queens at once. Frank Pellett's history of American beekeeping says Doolittle "gave out his new system of dipping artificial cells" in 1889 and treats him as the founder of commercial queen rearing.

    bushfarms.comchestofbooks.com

    Credit

    G. M. Doolittle

  2. 1919

    The Buckfast bee

    By Brother Adam's own account, Buckfast Abbey in England had 46 colonies in 1915, and only 16 survived the next winter after the Isle of Wight disease, a plague then blamed on the acarine mite; all the survivors were Ligurian, an Italian strain, or of Ligurian origin. He took over the abbey's bees in 1919, and that year raised a queen, registered as breeder B-1, that he called "the progenitor of the present-day strain." He went on to cross bees of many races, mating queens at an isolated apiary on Dartmoor, and traveled "more than 100,000 miles" to collect them; the abbey says Buckfast queens are now kept by beekeepers worldwide.

    LessonThe stock that survived an epidemic became the base for breeding.

    bibliography.karlkehrle.orgbuckfast.org.uk

    Credit

    Brother Adam (Karl Kehrle), Buckfast Abbey

  3. 1920s

    Queens inseminated by instrument

    A queen mates in flight with many drones (males), so a breeder cannot normally choose her mates. A 2024 review in the journal Hereditas credits Lloyd Watson as the first, in the 1920s, to show reproducible success at placing drone semen in a queen by instrument; he mostly used one drone per queen, too little semen for a full-strength colony. Harry Laidlaw, later a professor at the University of California, Davis, showed in work published in 1944 that the valve fold, a flap inside the queen, must be held away from the oviduct opening, and built the first functional instrument. O. Mackensen's carbon dioxide anesthesia (1947) and better instruments followed, and the review calls the method "reliable" today.

    pmc.ncbi.nlm.nih.govbeeculture.com

    Credit

    Lloyd R. Watson; Harry H. Laidlaw Jr.; O. Mackensen

  4. 1957

    African queens escape

    Biologist Warwick Kerr was crossing European bees with bees from southern Africa in Brazil, to breed a strain suited to the tropics. In 1957, the Smithsonian reports, a replacement beekeeper accidentally released 26 African queens (Apis mellifera scutellata) near Rio Claro, São Paulo state. They mated with local bees, and the "Africanized" bees spread north through the Americas, reaching Mexico's Yucatán Peninsula in 1989.

    LessonBreeding stock that escapes cannot be recalled; one release changed how bees are kept across the Americas.

    torch.si.edu

    Credit

    a failure, in Warwick Kerr's breeding project

  5. 1990s

    Breeding values for bees

    Livestock breeders rank animals by a breeding value, an estimate of the genetic merit they pass on, computed with a statistical method called BLUP (best linear unbiased prediction). BLUP was adapted to bees, where a colony's results reflect both its queen and her workers, and from the mid 1990s it ran on computerized records, now kept in the shared online service BeeBreed. Across 195,458 tested queens of the Carniolan bee in 12 countries, the genetic gain in honey yield was about 2 kg from 1960 to 1985, then over 5 kg from 1992 to 2018, without higher inbreeding, by the analysis of the institute that runs the evaluation.

    pmc.ncbi.nlm.nih.gov

    Credit

    Kaspar Bienefeld and colleagues, Institute for Bee Research Hohen Neuendorf, Germany

  6. 1996

    Mite resistance, first misnamed

    In 1996 John Harbo of the USDA and Roger Hoopingarner of Michigan State University found a few colonies in which varroa mites multiplied slowly, apparently because many mites failed to breed, and named the trait "suppressed mite reproduction." Later work showed that worker bees were finding and removing brood cells with breeding mites inside, so the trait was renamed varroa sensitive hygiene (VSH), and breeding stock went to the industry from 2001 through Glenn Apiaries. In a 2004–2006 test run with Alabama beekeepers, 12% of VSH colonies, 24% of Russian colonies (a USDA stock from eastern Russia) and 40% of Italian colonies needed treatment against varroa, with similar honey yields.

    LessonThe first name described what seemed to happen, not what the bees did.

    ars.usda.govars.usda.gov

    Credit

    John Harbo, Roger Hoopingarner, Jeffrey Harris and Robert Danka (USDA-ARS, Baton Rouge); field test by Kenneth Ward, Robert Danka and Rufina Ward (Alabama A&M University)

  7. 2006

    The honeybee genome

    An international consortium published the honeybee's genome, its full DNA sequence, in Nature in October 2006. Compared with other sequenced insects it found fewer genes for immunity and for breaking down toxins, more genes for smell receptors, and population genetics that pointed to an African origin for the species. The sequence is the reference that later DNA-based breeding work relies on.

    pmc.ncbi.nlm.nih.gov

    Credit

    Honeybee Genome Sequencing Consortium

Watch for

  1. Californian hybrids resist mites

    A University of California, Riverside team monitored 236 colonies over four years and reports that colonies of a semi-feral Southern California hybrid (a mix of Western European, Eastern European, Middle Eastern and African lineages) carried consistently fewer varroa mites than colonies with commercial queens, crossed the treatment threshold of 3 mites per 100 bees less often, and had larvae that mites found less attractive in lab tests. A June 2026 report puts the difference at 68% fewer mites on average and quotes the lead author that the colonies' temper "can be a lottery." A companion survey by the same group found no significant difference in parasite levels that beekeepers reported for the two stocks.

    Achieved whenanother group measures lower mite loads in this stock at another site with the same methods, and scores its defensiveness.

    pmc.ncbi.nlm.nih.govambrook.com

    Claimed by

    Genesis Chong-Echavez, Boris Baer and colleagues, UC Riverside, March 2026

Open

  1. 01

    Camera-guided queen insemination

    A camera on the insemination microscope, with software that finds the queen's sting chamber, the valve fold and the syringe tip in the live image and shows a beginner where to move, or drives a motorized stage for the final step of passing the tip beyond the valve fold.

    The brief
    Credit · unclaimed
  2. 02

    Score mite breeding by camera

    A camera on the stereo microscope used to score varroa reproduction, with software that recognizes the bee pupa's age and the mite offspring in each opened brood cell and keeps the tally, plus the public labeled image set it is trained on. Breeders use this score, the share of mites that fail to breed, to pick colonies for mite resistance.

    The brief
    Credit · unclaimed

Open 01 · the brief

Camera-guided queen insemination

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A camera on the insemination microscope, with software that finds the queen's sting chamber, the valve fold and the syringe tip in the live image and shows a beginner where to move, or drives a motorized stage for the final step of passing the tip beyond the valve fold.

Why it is still open

The University of Florida's entomology department described instrumental insemination in March 2026 as "tedious" work that "takes specialized training and precision instruments", in which the operator lifts the valve fold with the syringe tip by hand. Searches of the web, Europe PMC and Crossref in November 2026 found manual instruments only, the newest an improved manual device described in Bee World in 2025 (Van Laere and de Graaf), and no camera-guided or robotic insemination of queens.

blogs.ifas.ufl.edu

Built with what exists

A standard insemination instrument and stereo microscope with a camera port; motorized micromanipulator stages of the kind used for injecting cells; an object-detection model such as Ultralytics YOLO trained on labeled microscope video; carbon dioxide anesthesia, as in the standard method.

  1. First step

    Record and label microscope video of thirty inseminations by an experienced operator, marking the valve fold and syringe tip frame by frame.

  2. Done looks like

    Fifty queens inseminated by first-time operators using the guide and fifty by the same operators without it, with a published tally of queens laying worker brood three weeks later, and minutes per queen.

Open 02 · the brief

Score mite breeding by camera

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A camera on the stereo microscope used to score varroa reproduction, with software that recognizes the bee pupa's age and the mite offspring in each opened brood cell and keeps the tally, plus the public labeled image set it is trained on. Breeders use this score, the share of mites that fail to breed, to pick colonies for mite resistance.

Why it is still open

In a study across 13 European countries, Fanny Mondet and colleagues (Insects, 2020) write that judging the stages of mite offspring "requires a considerable level of knowledge, skill, and experience", that at least 35 cells each holding a single mite must be opened under a stereo microscope to score one colony reliably, and that the work is "difficult and labor intensive"; they trained scorers with a shared photo guide and a workshop. The image models published for varroa count mites on sticky boards or on bees, and searches found none that scores mite offspring in opened cells.

pmc.ncbi.nlm.nih.govzaguan.unizar.es

Built with what exists

A stereo microscope with a camera port and a USB microscope camera; the photo guide of developmental stages used in the European study; Label Studio for labeling; Ultralytics YOLO or a small open vision model fine-tuned on one graphics card (see the Note "One GPU to fine-tune").

  1. First step

    Ask a laboratory that keeps frozen brood frames for mite-reproduction scoring (the European study stored them at −20 °C for up to six months) to film each cell it opens through the microscope, with the scorer's call written beside it.

  2. Done looks like

    At least 2,000 labeled images of opened infested cells from at least three laboratories, published under CC BY; on cells from a laboratory not used in training, the tool's call of breeding or not breeding agrees with an expert scorer on at least 90% of cells, with the agreement between two experts reported beside it, and minutes per colony compared with scoring by hand.