Sugar shake tests, or alcohol washes; what’s your method? Until a couple of weeks ago, I thought that sugar shake/roll tests to estimate Varroa destructor (common name varroa) levels in a colony were benign and caused no harm to the bees— apart from a bit of shaking and annoyance. Now I know that they do harm bees.
A few years ago, I carried out a small research project to test the sugar shake tests against alcohol washes for accuracy. I found no significant difference so I carried on with sugar shakes for a couple of years because I thought it was better not to kill even 300 bees. Then I realised that alcohol wash tests were quicker, less messy and didn’t encourage wasps or promote robbing. It was difficult keeping the collection bucket dry with sugar shakes because some of the foragers regurgitated nectar and the bucket got sticky quickly and had to be washed and dried between colonies. When I stopped, there was no icing/powdered sugar around the hives afterwards to entice robbers anymore either so that was good. I haven’t done a sugar shake test for three years.
Bees Disappear
There’s an edited version of my project at the end if you want to read how I tested the methods. The original article was published in BeeCraft magazine. The collection method of bees for both tests is the same, but sugar shake tests are history now. This is because new research by Bruckner et al., (2025)1 shows that 56% of the bees from sugar shaken samples disappear from the hives within five days of testing. So, what happens to them?
Chemical Disruption
Bateson et al., (2011)2 studied honey bee cognition in relation to negative experiences. It has been known for some time that stressed vertebrates, such as animals and birds with poor welfare, exhibit impaired cognition and decision making. They experience something like pessimism and expect the worst to happen. However, it was not known how invertebrates might respond to negative experiences until the recent Bruckner experiment. Before the bees were exposed to vigorous shaking they were conditioned and taught to respond to expect a reward after one stimulus, and a punishment after another stimulus in the same sensory area. In this situation it was a response to different odours and measured by whether the bee extended its proboscis or not.
It turns out that shaken bees have lower levels of haemolymph, dopamine, serotonin and octopamine. The latter is a neurotransmitter that functions during reward learning. In Bateson’s experiments, they concluded that shaking bees causes them to have an increased expectation of punishment, pessimism and a negative outcome, and impaired cognition. Shaken bees may not be able to remember how to get back to their hives. The Bateson experiment also suggests that honey bees could be considered to be capable of exhibiting emotions.
Field Studies
In the Bruckner field experiments 20 full sized colonies were used. 900 worker bees from each colony were chilled and marked with coloured Sharpie pens. There were three treatment groups; powdered sugar shake, powdered sugar coating only, and the control group which was marked only and kept in the shade to recover. 300 bees were sampled in each test. After five days 44% of the powdered shaken bees were recaptured compared to 72% of the sugar dusted group, and 76% of the control group. Not surprisingly, beekeepers are now advised to use alcohol washes rather than sugar shake tests to estimate varroa infestations.
Estimating Varroa Levels
Eight years ago, I was interested in evaluating Véto-pharma’s (now Vita (Europe)) product Varroa easyCheck, and comparing the sugar shake and alcohol test methods of assessing varroa levels in a colony. Out of curiosity, I carried out a small research project.
Monitoring Varroa
Monitoring natural the varroa drop through an open mesh floor (OMF) onto a sticky board is a useful technique. You can measure drop over a 7- day period, then calculate the average daily drop. This is a common method because is easy to carry out and doesn’t involve disturbing or manipulating a colony of honey bees. However, Oliver3 suggests that natural mite drop can be greatly under or overestimated, and may vary widely from day to day in the same hive.
In the past, I have used the BeeBase varroa calculator4 which advises when treatment should be carried out based on daily natural mite drop results. It takes into account the time of year and strength of the colony since 20 mites dropping from a 5- framed nucleus is more serious than the same number falling from a 20-framed double brood box. However, recent work by Professor Stephen Martin suggests that the NBU threshold level of harm is too low, and many beekeepers over- treat colonies for varroa as a result. I now refer to Professor Martin’s calculator to inform my own management: https://www.varroaresistant.uk/mite-calculator/
My experience has shown that unless the sampling board is sticky from a product like Vaseline live varroa can escape, or dead varroa blow away. If the open mesh floor is clogged with debris, it might prevent the varroa dropping through which skews results. In the past, some of my hives didn’t have convenient trays to hold the sticky boards so they were held in place below the hives. I noticed great tits and blackbirds feeding on these boards. They may have been eating dropped pollen and dead bees, but perhaps varroa also. Needless to say, I now have well-fitting varroa boards.
The methods that I’m about to describe are probably more accurate than counting a natural mite drop if you are looking for a snapshot of varroa levels on a given day, or you want to cross reference with natural drop results. It only takes a few minutes to give you a reasonably accurate assessment of a colony’s varroa infestation levels.
It was previously believed that varroa compromise honey bee immunity and health by feeding on the honey bees’ haemolymph and thereby transmitting viruses. However, Samuel Ramsey, whilst a PhD student at the University of Maryland (USA) discovered that varroa feed mainly on the bees’ fat bodies, which is more serious than we previously thought and also compromises bee nutrition. We know now that feeding depends on varroa life stage and that the immature mites in the cell feed largely on pupal haemolymph.
Fat bodies are specialised organs that function in a similar way to our livers. They are suspended in the bee’s haemocel (abdominal body cavity) and have immediate contact with the bees’ haemolymph (blood). They are a bee’s principal organ of metabolism, functioning in many aspects of a bee’s physiology, including energy storage, detoxification, and synthesis of important proteins called vitellogenins (responsible for stimulating egg production). They are also essential in maintaining a healthy immune system. Varroa infestations compromise all these areas rendering bees vulnerable to infection and ill health. Bee health can be further compromised by careless beekeeping, poor weather and lack of forage, and exposure to pesticides—both agricultural and beekeeping chemicals.
Sugar Shake v Alcohol Wash
With so many new beekeeping products appearing on the market, I was curious to see if Veto- pharma’s Varroa easyCheck alcohol wash test was more accurate than a sugar shake test, so I tested 7 colonies in the summer of 2017 using both methods on each colony at the same time.
The test involved collecting two samples of 300 live bees from the brood boxes of each of the seven test colonies. I wanted to collect young nurse bees so I chose frames that had a lot of unsealed brood. I shook each chosen frame into a large, dry plastic bucket allowing the flying bees to leave before I swirled it around to keep the remaining young bees at the bottom while I scooped up 300 of them in a measuring container—a half cup measure used in baking. This is the volume of 300 bees.
The easyCheck Test
The easyCheck container has 2 measuring lines inside the removable basket. The lower mark measures 200 bees, and the upper mark 300. The instructions suggest running the mesh basket up the frame to collect and remove bees but, having tried this, I found it easier and quicker to and obtain all my samples by shaking bees into the bucket first and then scooping up a half cup of bees.
This method involves adding alcohol to the test container, thereby killing the sample of bees. 5% alcohol (i.e., winter-strength car windscreen washer fluid) is recommended by Véto-pharma, so this is what I used.
Following instructions, I poured fluid into the test container up to the bottom line marked on the outside. This line was difficult to see so I enhanced it by tracing a black marker pen over it. Next, I poured in the sample of 300 bees and then I inverted the container, swirling it to kill the bees quickly. At this point, I opened container again and topped up the fluid to the top marker line. I then swirled the bees round in the container for 60 seconds using my stop-watch to time this. Now the mites could be seen at the bottom of the container. Finally, the fluid was strained through mesh and reused, and the varroa count was confirmed.
The Sugar Shake Method
I collected my sample of 300 bees, as above, and placed it in a one-pint glass jar fitted with a mesh lid. I swirled the jar around to keep the bees on its bottom as I opened the lid to insert 2 heaped tablespoons of icing sugar. I rolled the jar around a few times cover the bees in icing sugar. I left them in the shade for two minutes to get all hot and bothered inside the jar. This made the varroa too hot, so they crawled out from under the bees’ tergites. Incidentally, if you see varroa on bees in your hive, you can be sure that there is a heavy infestation because the varroa are usually hidden away in the abdominal area feasting on fat bodies.
I then shook the jar onto a plate (you need a solid plate as disposable ones can blow away at the crucial moment—believe me) so that the icing sugar and the mites fell out onto the plate. I shook each sample for the recommended 90 seconds and noticed that the mites usually started falling after the first minute.
The sugar-dusted bees were then dropped out onto the top bars of their hive. They walked down into the frames fairly quickly. I was not stung during the study.
Using the water spray, I sprayed the white sugar on the plate to reveal the dark varroa mites which I then counted before cleaning plate for next test.
The Results
| Colony | Mite count: Sugar Shake | Mite count: Alcohol Wash |
| 1 | 2 | 1 |
| 2 | 4 | 3 |
| 3 | 2 | 1 |
| 4 | 3 | 1 |
| 5 | 4 | 6 |
| 6 | 16 | 6 |
| 7 | 0 | 0 |
| Mean | 4.4 | 2.6 |
A Student’s paired t-test was performed and it was found that P=0.42
Note: P = the probability of getting the difference observed by chance. If P< 0.05 then this would indicate that the difference found was significant (not due simply to chance).
Discussion
The small difference in the average mite counts between the two test methods evidently arose by chance, such as by shaking bees from the same hive. Therefore, I found no statistically significant difference between these two methods for testing the varroa level in a colony. Nevertheless, the two methods have some significant differences in how they affect the bees and the beekeeper.
With the sugar shake method, the bees are alive at the end of the test, which I like. However, this method takes longer to perform than the alcohol wash, and it is messier and stickier.
With the easyCheck alcohol wash method, the bees die, but it is quicker, less messy, and it requires less equipment.
While I shall probably continue to use the sugar shake test on my own few colonies, bee farmers and others may prefer using the quicker easyCheck test. Some beekeepers may find that they can easily use the mesh basket to collect bees and this will make the test even easier for them to perform.
However, if you capture the queen by mistake you will be sorry.
(3) Treatment Threshold: If there are over 2% varroa in your sample of 300 bees then treatment is required before August. In August treatment is recommended above 3%.
Please note, sugar shake tests are not recommended now. Treatment-free beekeepers, and those working towards this, probably monitor varroa intensively by counting daily natural mite drop. This is because they want to look for evidence of hygienic behaviours which might manifest in chewed-up pupae, immature varroa mites, or mites with missing legs. I monitor mite drops daily and do alcohol wash tests during the active season if I want to confirm results.
References:
1Bruckner, S., Williams, G. R., Tsuruda, J., & Underwood, R. M. (2025). Let’s not sugar coat it: the powdered sugar shake is not harmless for honey bee workers. Journal of Apicultural Research, 1–6. https://doi.org/10.1080/00218839.2025.2550855
2http://dx.doi.org/10.1016/j.cub.2011.05.017
3Oliver, R. Re-evaluating Varroa Monitoring, Americn Bee Journal, March 2020.
4BeeBase (National Bee Unit) website, www.nationalbeeunit.com
5ScientificBeekeeping.com
Discover more from The Beelistener
Subscribe to get the latest posts sent to your email.
Hello Ann. Interesting article. I note that you are using Whitworths icing sugar which has an anti caking agent (tricalcium phosphate). I am not sure this does the bees much good and would be interested to know what brand Bruckner used as most brands include some sort of anti caking agent.
Hello Marina,
Thank you for reading the article and commenting. They don’t say what brand they used, or whether anti-caking agents were involved. Since the damage was caused by shaking I don’t suppose it makes any difference to the experiment. Best wishes, Ann.
Thank you for this Ann. It came at exactly the right moment as I’m preparing my lesson on varroa for my forthcoming Module 3 Study Group.
I’d been uncomfortable for some time about the sugar roll being harmless to the bees but didn’t want to go against the received wisdom when preparing candidates for exams.
Hello, Alan. Thank you for your comments and appreciation of the blog post. I feel somewhat bad about not wondering about the effect upon the bees before now. I guess some people are better about thinking outside the box than others. I am delighted that this helps with your teaching programme because passing on information and helping people become more curious is my passion.
Best wishes, Ann.
Hello Ann, Thank you as always for a very interesting article. Like you, I preferred the sugar shake rather than killing the sample with alcohol, but found the sugar easily got sticky, so I have been using CO2 to anesthetise them. Have you ever tried this and do you know of any studies that have considered the bees survival after this and its efficacy compared with an alcohol wash?
Regards, David
Hello, David. Thank you for commenting. I have tried using CO2 but didn’t get along well with the kit. I found that the gas canisters released all the gas at once when I removed them from the container of bees. I also noticed a lot of moisture in the container so I was not confident that the varroa actually came off the bees.
I haven’t seen any studies on this method but I haven’t been searching. You could try Google Scholar or Pubmed. You could even contact the authors of the Bruckner paper.
They used vigorous shaking in the experiments, as you would with sugar shake tests; the CO2 method advises gentle shaking. But are the bees affected by vigorous shaking if they are anaesthetised? Is the CO2 method harmless? Another study is needed to find that out I would say, but even with the sugar shake test some of the bees survive whereas they all die with alcohol washes. If you consider that bees may have more sophisticated brains that are capable of emotions it adds another dimension to decision making.
I count the daily varroa drop from all five colonies so have not even done alcohol wash tests over the last year. By examining the varroa under a magnifying glass I have been able to also count immature and damaged mites and chewed honey bee pupae in my quest to identify colonies that can keep varroa levels low by themselves.
I digress, but, re treatment equipment, I have tried using the Gasvap kit but that fell apart at the crucial moment. I believe that the kit has been modified to correct the problem but I haven’t had great experiences with sublimation either. Varrox Eddy has such a thick wand it didn’t fit in the entrances of most of my hives.
Best wishes, Ann.