0. Abstract

Sleep appears to play an important role in the lives of honey bees, but to understand how and why, it is essential to accurately identify sleep, and to know when and where it occurs. Viewing normally obscured honey bees in their nests would be necessary to calculate the total quantity and quality of sleep and sleep’s relevance to the health and dynamics of a honey bee and its colony. Western honey bees (Apis mellifera) spend much of their time inside cells, and are visible only by the tips of their abdomens when viewed through the walls of an observation hive, or on frames pulled from a typical beehive. Prior studies have suggested that honey bees spend some of their time inside cells resting or sleeping, with ventilatory movements of the abdomen serving as a telltale sign distinguishing sleep from other behaviors. Bouts of abdominal pulses broken by extended pauses (discontinuous ventilation) in an otherwise relatively immobile bee appears to indicate sleep. Can viewing the tips of abdomens consistently and predictably indicate what is happening with the rest of a bee’s body when inserted deep inside a honeycomb cell? To distinguish a sleeping bee from a bee maintaining cells, eating, or heating developing brood, we used a miniature observation hive with slices of honeycomb turned in cross-section, and filmed the exposed cells with an infrared-sensitive video camera and a thermal camera. Thermal imaging helped us identify heating bees, but simply observing ventilatory movements, as well as larger motions of the posterior tip of a bee’s abdomen was sufficient to noninvasively and predictably distinguish heating and sleeping inside comb cells. Neither behavior is associated with large motions of the abdomen, but heating demands continuous (vs. discontinuous) ventilatory pulsing. Among the four behaviors observed inside cells, sleeping constituted 16.9% of observations. Accuracy of identifying sleep when restricted to viewing only the tip of an abdomen was 86.6%, and heating was 73.0%. Monitoring abdominal movements of honey bees offers anyone with a view of honeycomb the ability to more fully monitor when and where behaviors of interest are exhibited in a bustling nest.

1. Background and Research Question

This article examines a behaviour still poorly documented in the honey bee: sleep inside honeycomb cells. The authors begin from the observation that bees spend a considerable part of their lives in these cells, where they store honey and pollen, rear brood, contribute to the colony's thermal regulation, and probably also sleep.

The primary objective is to determine whether observation of the abdominal tip — the only part generally visible when a bee is deeply inserted in a cell — allows reliable identification of sleep and its distinction from other behaviours such as cell tending, feeding, or brood heating.

The study draws on the premise that sleep in bees can be recognised by relative immobility combined with discontinuous ventilation, that is, series of abdominal pulsations separated by long pauses.

2. Method

Figure 1: Observation hive with honeycomb slices and exposed cells. (A) Still frame from an infrared video of an occupied hive, with the hive entrance leading to a tubular tunnel at the lower left. (B) The comb slices were held in place with nails inserted into the wooden frame and aligned at natural distances from one another. (C) The comb on the right was tilted before the study to show its width and part of the cell contents, including capped and open brood. The lower rear corner of each comb was removed to facilitate movement of workers within the hive. Photos by Barrett Klein.

The authors used a small colony housed in a specially constructed observation hive with comb slices presented in cross-section, making the interior of certain cells visible. Behaviours were filmed using an infrared camera and a thermal imaging camera.

Four behavioural categories were defined: sleeping, cell tending (including cleaning or comb-building activity), feeding, and heating. Sleep was identified by discontinuous ventilation and marked immobility. Heating was identified by continuous ventilation and comparable immobility, but with a thorax markedly warmer than the immediate surroundings.

The researchers conducted 49 sampling sessions over 34.5 hours, observing 115 visible cells. They then analysed a subset of bees to describe ventilatory pulsations in detail and to compare thoracic temperature with that of the surroundings. Finally, a test was conducted with 54 human observers, who were asked to identify behaviours from videos in which only the abdominal tip remained visible.

3. Main Results

Figure 3: Still frames from infrared videos showing bee behaviours, all head-first inside cells. (A) The sleeping bee, at the centre of the cell, is oriented to the left, ventral side up. (B) The feeding bee has its mouthparts extended and its body less fully inserted into the cell; it is oriented to the left, ventral side down. (C) The heating bee is oriented to the left, ventral side facing the observer (lateral view). (D) The sleeping bee in the centre is oriented to the left, ventral side down, and should be compared with the heating bee on the right, oriented to the right, dorsal side facing the observer (lateral view). All other bees inside cells are tending (cleaning or building) cells, including the two visible in (A) and the three in (B), indicated by black circles superimposed over the centre of their thorax. Contrast and brightness adjustments serve to highlight the sleeping, feeding, and heating bees. Images by Barrett Klein.

Of 455 behavioural events observed inside cells, 76.4% involved cell tending, 16.9% sleep, 6.4% heating, and 0.4% feeding. Sleep thus emerges as the second most frequently observed behaviour inside cells.

Sleep episodes were characterised by sequences of abdominal pulsations separated by prolonged pauses, often exceeding 10 seconds. Heating bees, by contrast, showed continuous ventilation. The short inter-pulse intervals were comparable across behaviours, but the presence or absence of long pauses clearly distinguished sleeping from heating.

Thermal measurements show that only heating bees had a significantly higher thoracic temperature than their surroundings. Sleeping bees, those engaged in cell tending, and those feeding showed no clear thermal difference from their immediate environment.

When visibility was deliberately restricted to the abdominal tip, observers correctly identified sleep in 86.6% of cases and heating in 73.0% of cases. Cell-tending and feeding behaviours were more often confused with one another.

Behaviours observed inside cells were recorded during both day and night, with no marked difference between periods for sleep, cell tending, or heating in this experimental setup.

Video 1. Infrared video of a bee sleeping inside a cell. The bee in the centre is oriented to the left, ventral side up. Note the worker bees tending (cleaning or building) other cells.

4. Interpretation and Limitations

Figure 9: Still frames from thermal imaging videos. (A) A worker heating inside a cell, head oriented to the left, with part of the abdomen, wings, and hind legs protruding from the cell to the right. (B) The two brightest spots on the left each show the relatively warm thorax of a bee heating inside a cell: one immediately behind the plastic window of the hive (upper arrow), and another at a depth of one cell, visible through the wax wall of a cell (lower arrow). These workers spent approximately 5 minutes and 25 minutes respectively heating inside these cells. (C) The arrows indicate a heating bee whose body temperature fluctuates. The total heating time exceeded 11 minutes. (D) Each bright spot corresponds to a relatively warm thorax, but of a cell-tending bee, not a heating one.

The authors conclude that non-invasive observation of the ventilatory movements of the abdomen allows reliable detection of sleep in honeycomb cells. This approach is particularly useful in situations where the bee is almost entirely concealed.

The study also reinforces the view that young house bees preferentially sleep in cells, probably more so than older foragers. Cells may offer a relatively disturbance-protected location, a favourable microenvironment, or simply a space used between several tasks near the brood nest.

The authors nevertheless highlight several limitations: the colony studied was small, the experimental hive differed from a standard hive, some peripheral cells had been emptied to improve visibility, and these modifications may have influenced the relative frequency of the observed behaviours, in particular cell tending, heating, and sleeping.

They therefore caution that the proportions measured should not be generalised without care, even if the behavioural criteria for distinguishing sleep from heating appear robust.

5. Conclusion

This study demonstrates that honeycomb cells do not serve exclusively for storage and brood rearing: they also constitute an important space for adult behaviours, particularly sleeping and heating.

In a bee inserted into a cell, the combination of two visible cues — the type of abdominal ventilation and the absence of large body movements — is in most cases sufficient to distinguish a sleeping bee from a heating one. This opens the way to less invasive observation of sleep within the colony and to a better understanding of its function in the social organisation of bees.

► Read the article in English

---

See also:

• Sleep in honey bees (Apis mellifera)
• Bee behaviour inside the hive: insights from long-term video analysis
• PDF, regulator of the honey bee's biological clock
• Hive insulation put to the test of collective thermoregulation in bees
• The inner life of insects

 

Source Reference

Klein, B. A., & Busby, M. K. (2020). Slumber in a cell: honeycomb used by honey bees for food, brood, heating… and sleeping. PeerJ, 8, e9583. DOI: 10.7717/peerj.9583.