Feeding and Checking Bees Before Winter

If a colony is going to die this winter, the decision was made in August. Purdue Extension explains why: the varroa mite population in the hive usually peaks in early fall, at the same moment the colony is raising the workers that have to survive until spring. Those winter bees live up to six months, against about six weeks for a summer bee, according to Penn State Extension. They are a different kind of bee, and a mite population feeding on them at the wrong moment takes the whole colony down in January.

The mechanism got rewritten in 2019. Ramsey and colleagues, in a paper hosted by USDA ARS, found that varroa feed primarily on fat body tissue rather than hemolymph. About 72 percent of the signal in their biostain work matched fat body against 17 percent for hemolymph, and mites fed only hemolymph survived 1.8 days with 5 percent egg production, against 3.5 days and 40 percent on fat body. The fat body is where vitellogenin comes from, and vitellogenin is what a winter bee runs on. Varroa do not drink bee blood. They eat the organ that stores winter.

Treating in fall is treating late

The common phrasing is fall treatment, and every source that gives a deadline puts it earlier than that. Cornell’s overwintering guidance is the bluntest: treating bees for the first time in September or October is already too late. University of Minnesota Bee Lab says treat before mid-August to protect developing winter bees. Penn State frames it as treating ahead of the winter bee generation, and says fall mite control is a major factor linked to overwintering survival. NC State adds the line that explains a lot of dead-outs: many colonies that succumb to varroa do so in late summer or fall, not in the cold.

Count them properly

A half cup of bees is the standard sample, and it is roughly 300 bees according to the Honey Bee Health Coalition, NC State, Minnesota, and Purdue. Mississippi State calls the same half cup 400 bees. That difference is not academic. The same raw mite count produces a 33 percent different infestation rate depending on which conversion gets used.

Sample from one to three brood combs, queen excluded. Penn State calls the alcohol wash the most accurate method for monitoring mite populations, and Mississippi State says it is much more accurate than the sugar shake. Minnesota gives the reason: fewer variables affect the alcohol wash, where humidity and technique degrade the sugar roll. The wash kills the sample. Three hundred bees against a colony’s winter is a trade worth making.

Two methods carry less weight than their popularity suggests. The Honey Bee Health Coalition puts ether roll detection at only 50 to 60 percent of mites present, and puts sticky boards in its lower-confidence category, hard to read as a population estimate. The published thresholds bear that out: 10 mites a day at the Coalition, over 30 at USDA ARS, 50 at Purdue, 60 to 190 at NC State. A nineteenfold spread across credible institutions is itself the argument against the method.

Screened bottom boards are the other overrated tool. The Coalition rates them least effective, perhaps 5 to 10 percent effective and only in northern states. Penn State says mite loads often still reach economic thresholds in hives that have them.

The threshold nobody agrees on

Published action thresholds for late season run from under 1 percent to 10 percent, which is a wide enough band that the choice matters more than the count.

• University of Illinois Extension: less than 1 mite per 100 bees is acceptable in late fall.
• Penn State: keep levels below or around 2 mites per 100 bees.
• Minnesota Bee Lab: below 1 percent is optimal, treat above 2 percent, which is 6 or more mites in a 300-bee sample.
• USDA ARS and NC State: 3 mites per 100 bees.
• Cornell’s Northeast calendar: 2 per 100 in August, 3 per 100 in September and October.
• Purdue: 5 to 10 percent late season, which is 15 to 30 mites per 300 bees.

Purdue’s late-season figure sits an order of magnitude above Illinois and well outside the Coalition’s 1 to 3 percent action band. Treat it as the outlier it is. The Coalition released a ninth edition of its varroa guide in June 2026 and the numbers above come from the eighth, so anyone working to a threshold should check the current edition rather than a blog. On products, the Coalition’s language is worth quoting directly: using any registered product in a manner inconsistent with its labelling is a violation of federal and state laws.

Weigh the hive

Feed stores are the second half of the problem and they run on a real regional gradient. NC Cooperative Extension in the western North Carolina high country says 40 to 50 pounds, about one medium super. Mississippi State says 60 to 65 pounds. Cornell says 80 pounds for a full colony and 50 for a nuc in New York. Michigan State says 70 to 100. Illinois says at least 90 pounds, or a full deep box, or two medium supers.

Weight predicts survival better than pedigree does. A Penn State field study of 60 colonies, published by Döke and colleagues in 2019 and reported by the Entomological Society of America, found colonies under about 44 pounds net weight overwintered poorly, while those over about 66 pounds survived at roughly 94 percent. The same study found stock origin uncorrelated with survival, across two southern and two northern stocks. Southern bees do not fail in the north because they are southern. Light colonies fail because they are light.

Syrup has a temperature cutoff, not a date

Fall syrup is 2 parts sugar to 1 part water, roughly 66 to 67 percent sugar, at Michigan State, Cornell, Illinois, Georgia, and UF IFAS. Nobody found recommends going thicker. Cornell says a 3:1 mix causes crystallization problems.

The stopping point is where it gets specific. Illinois says the window closes when temperatures fall below 50°F. Oregon State says stop below 57°F, while noting the Honey Bee Health Coalition’s 50°F figure. The reason is the same in both cases: bees have to evaporate water out of syrup to store it, and they cannot do that in a cold hive. Mississippi State makes the same point about chilling bees with cold syrup. Calendar cutoffs follow the same gradient as everything else, October 1 in Michigan, end of September in New York, end of October in Mississippi. Cornell also puts the last pollen or pollen substitute feeding in September.

Moisture, and the thing everyone repeats about it

Shivering bees produce metabolic water all winter, and Gary Reuter and Marla Spivak at the University of Minnesota describe the job as keeping it from freezing inside the colony. Their answer is an air path over a moisture board, three-quarter inch fiberboard or a folded pile of newspaper, with the outer cover left crooked so air reaches it. No plastic, no styrofoam. Reuter’s upper entrance is a hole half an inch to an inch across, three inches above the bottom edge of a deep body.

The claim that moisture kills more colonies than cold does is repeated everywhere and could not be sourced to any extension service or USDA publication. The mechanism is documented. The ranking against cold is not.

There is also a controlled trial pointing the other way. St. Clair, Beach and Dolezal at the University of Illinois, publishing in PLOS ONE in 2022, wrapped colonies in central Illinois and got 4.8 percent winter mortality against 27.3 percent in uncovered controls, with covered colonies losing mass at roughly half the rate. That benefit came from retaining heat, not from venting. No source found resolves that tension with numbers, so a beekeeper choosing between more ventilation and more insulation is making a judgment call rather than following the research.

Combine the weak ones now rather than watching them fail. Mississippi State says hopelessly weak colonies should be combined over a sheet of newspaper so the odors mix and the fighting stays down, and that diseased colonies should never be combined. UF IFAS puts an adequate winter colony at 25,000 to 30,000 bees and says colonies under 15,000 are not likely to survive freezing temperatures. Two mediocre colonies going into November are one decent colony in March, or two dead ones.

Sources

• Ramsey et al. 2019, PNAS, hosted by USDA ARS — https://www.ars.usda.gov/ARSUserFiles/333/Bauchan/Varroa%20destructor%20PNAS%20%20Ramsey%202019.pdf
• Purdue University Extension, POL-8 — https://extension.entm.purdue.edu/publications/POL-8/POL-8.html
• Penn State Extension, Honey Bee Management Throughout the Seasons — https://extension.psu.edu/honey-bee-management-throughout-the-seasons
• Penn State Extension, Methods to Control Varroa Mites — https://extension.psu.edu/methods-to-control-varroa-mites-an-integrated-pest-management-approach
• Penn State Extension, Alcohol Wash for Varroa Mite Monitoring — https://extension.psu.edu/alcohol-wash-for-varroa-mite-monitoring
• University of Minnesota Bee Lab, Varroa Mite Testing — https://beelab.umn.edu/varroa-mite-testing
• Cornell CALS Pollinator Network, Overwintering — https://cals.cornell.edu/pollinator-network/beekeeping/overwintering
• Cornell CALS Pollinator Network, Beekeeping Calendar for the Northeast — https://cals.cornell.edu/pollinator-network/beekeeping/beekeeping-calendar-northeast
• USDA ARS Carl Hayden Bee Research Center, When to Sample — https://www.ars.usda.gov/pacific-west-area/tucson-az/carl-hayden-bee-research-center/research/varroa/how-to-monitor-sample-and-treat-for-varroa/when-to-sample/
• NC State Extension, Managing Varroa Mites in Honey Bee Colonies — https://content.ces.ncsu.edu/managing-varroa-mites-in-honey-bee-colonies
• Honey Bee Health Coalition, Tools for Varroa Management, 8th edition — https://honeybeehealthcoalition.org/wp-content/uploads/2022/08/HBHC-Guide_Varroa-Mgmt_8thEd-082422.pdf
• Honey Bee Health Coalition, ninth edition announcement — https://honeybeehealthcoalition.org/honey-bee-health-coalition-releases-ninth-edition-of-the-tools-for-varroa-management-guide/
• Mississippi State University Extension, Managing Varroa Mites in Honey Bee Colonies — https://extension.msstate.edu/publications/managing-varroa-mites-honey-bee-colonies
• Mississippi State University Extension, Colony Growth and Seasonal Management of Honey Bees — https://extension.msstate.edu/publications/colony-growth-and-seasonal-management-honey-bees
• University of Illinois Extension, Prepare Your Bees for Winter — https://extension.illinois.edu/blogs/know-how-know-more/2021-11-11-prepare-your-bees-winter
• Michigan State University Extension, Feeding Honey Bees (E-3369) — https://pollinators.msu.edu/sites/_pollinators/assets/File/FeedingHoneyBees-Final.pdf
• UF/IFAS Extension, ENY-161 — https://ask.ifas.ufl.edu/publication/IN1006/pdf
• University of Georgia Bee Program, A Synopsis of Bee Management in Georgia — https://bees.caes.uga.edu/beekeeping-resources/getting-started-topics/getting-started-a-synopsis-of-bee-management-in-georgia.html
• NC Cooperative Extension Watauga County, Fall Tips for Small Scale Beekeepers — https://watauga.ces.ncsu.edu/news/fall-tips-for-small-scale-beekeepers-in-the-high-country/
• Oregon State University Extension, Temps are dropping, should I change my bees’ diet? — https://extension.oregonstate.edu/ask-extension/featured/temps-are-dropping-should-i-change-diet-my-bees
• Reuter and Spivak, University of Minnesota, via eXtension Bee Health — https://bee-health.extension.org/wrapping-a-colony-for-a-northern-winter/
• Gary Reuter, University of Minnesota, Wintering Tips — https://garybees.cfans.umn.edu/wintering-tips
• St. Clair, Beach and Dolezal 2022, PLOS ONE — https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0266219
• Döke et al. 2019, reported by Entomological Society of America — https://entomologytoday.org/2019/01/09/colony-size-drives-honey-bees-overwinter-survival/