top of page
Peter Somers

Tricked or Treat: Epic Fail of the Sugar Shake


The unreliability of sticky boards and the dead bee effect of an alcohol wash has beekeepers doing the sugar shake. It works by tossing a half cup of bees (approx. 300) in a jar of powdered sugar to dislodge mites. Then you count them. This is supposed to give the beekeeper a fairly accurate estimate of the infestation level so a decision can be made on treatment. There’s been some indication, however, that the sugar roll (sugar shake, whatever) has failed to adequately warn beekeepers of an impending mite problem.

Step by step instructions detailing the method are widely available, particularly with the social media assist. A highly circulated instructional poster created by the University of Minnesota has been published or linked on many beekeeping sites, such as eXtension.org, the Bee Informed Partnership and Honey Bee Suite. Its widespread distribution is likely the result of clearly written instructions, helpful pictures, and practical treatment recommendations based on interpretation of the results:

If there is brood in the colony when you sample, you should double this [mite] number to factor in the amount of mites in worker brood. For example, if there are 5 mites / 100 bees, the total infestation is probably 10 mites/100 bees. If your colony has over 10-12 mites/100 bees, you should consider treatment.

Unfortunately, with every good intention, this particular U of M document is likely to establish your colony firmly on the road to disaster. Waiting for infestation levels to climb above 10% before issuing any recommendation to even consider treatment is a big mistake. Fifteen mites in the sugar jar ought to be a clear indication that treatment is overdue. For a colony of 40,000, it means 4000 mites in the hive! Colonies have collapsed with far less. For those that have relied on this particular publication to keep Varroa in check, it’s hard to imagine your bees are anything but dead right now.

Treatment parameters published by The Honey Bee Health Coalition are far less tolerant of the mite presence. Results are likely to be slightly more useful because some consideration is given to brood in the hive. Thresholds range from 1-5%. UDAF recommendations are in agreement, calling for a maximum infestation of 3% at any time. Using these lower limits, your colony is not certain to collapse before spring. But it certainly could. Enter Oliver.

Randy Oliver of scientificbeekeeping.com loves plotting data. The man has a total love affair with line graphs and bar charts. One of his more recent creations depicts the changing level of brood infestation in an untreated hive, as well as the corresponding results of an alcohol wash. Beginning the year with 100 mites, an August alcohol wash turns up just three, barely a 1% infestation. Two months later, the colony is overrun with 5,400 mites and suffers an October collapse.

The failure of the sugar shake/alcohol wash to detect a deadly mite infestation is largely a function of brood in the hive. There’s no way of knowing how many mites are under the brood caps, therefore there is no way of knowing the infestation level of the colony. Simply doubling the mite count is inadequate because there’s too many variables that affect the amount of brood in the hive: time of year, size of colony, available nutrition, etc. Sugar shake instructions that attempt to account for mites in the brood are structured around little more than a half decent guess. Whether you shake three mites or thirty, the fact that you shake any at all could mean there is little or no brood in the hive. But if there is brood, it could mean that brood is saturated with mites. Or not.

In other words, with barely any brood to invade, most mites will be attached to nurse bees (phoretic). That makes them easy targets for a sugar shake. Quite a few of them should end up in your jar. If you only get three, it must mean mite levels are low. Or you did something wrong. But, if you uncover three with lots of brood in the hive, you’re totally in the dark. You don’t know if mites are low, or if they’re all hiding in the brood. Without knowing, you’ve gotta treat.

Even if you shake zero mites! As much as that would appear your colony has Varroa under control, the reality is likely to be very different. But again… You can’t determine the infestation level of a colony without knowing the infestation level of the brood. Attempts at correcting for brood in the hive appear to be inadequate. On top of that, you may have used too small or large a sample, or not enough sugar, or shook the jar too lightly, or failed to set in the shade for a few minutes, or messed up the math. The sugar shake is so full of uncertainty, and so subject to user technique, that the margin of error is likely greater than the treatment thresholds.

So why bother? At best, you waste your time. At worst, the test gives you a false sense of security, so you do nothing and lose your bees. Undoubtedly, hives have been lost due to erroneous sugar shake results that advised against treatment. So, when it comes to the Varroa mite, don’t sugar coat it. Just treat.

bottom of page