Bee Buzz Box October 2026 What Bees See
Alan Wade and Peter Abbott
Canberra Region Beekeepers
Schlock and his one-time blind girlfriend.
Source: Wikipedia Schlock 1973.png https://en.wikipedia.org/wiki/Schlock_(film)
Is what bees see really so very different to what we see? The late bee vision researcher Adrian Horridge (1927-2024) tellls us that bees see the world entirely differently to vertebrates like ourselves. He provides us a succinct summary of what bees see (Horridge 2019a, Appendix).
‘I think that the bees suspect something!
’‘What sort of thing?’‘I don’t know. But something tells me that they’re suspicious!
’‘Perhaps they think that you’re after their honey.’‘It may be that.
You never can tell with bees.’
A. A. Milne, Winnie-the-Pooh (1926).
So just how do honey bee process the light signals they receive? Bees have no retina so their vision cannot be image based. Further bees have a brain far too small to process the ‘light of day’ in any sophisticated way. Adrian’s research has shown that bees see blue alone, use green receptors to adjudge position and polarised sky light to signpost direction.
With this information, we might ask ourselves how we can make sense of a visual system that is so foreign to ours. We can start by saying that honey bee sight must be governed by insect eye structure and by a rather different and simpler neural network (Horridge, 2019b). Since Adrian’s 19 April 2018 public oration to Canberra Region Beekeepers on how bees see things differently (Horridge, 2018), and the publication of his book on bee vision in 2019, we have spent pleasant hours with him pondering the deep mysteries of invertebrate vision. We have now attempted to distil these discussions to a layman’s version of the way honey bees actually see the world.
However, we first need to understand how Adrian came to his revolutionary notion of bee vision. To achieve this end he conducted a series of experiments where he trained bees to recognise simple patterns using sugar rewards. He then challenged these trained bees with new patterns to see what they had learnt. In so doing he abandoned the classical view that bees have trichromatic vision, see the world as we do and that they regard keepers of bees as a horde of barbarians after their treasure store.
To put it more simply, we have found that there is something intangible about the way bees see their world.
Turn-back-and-look
One good cue to explaining how bees see is signalled by their behaviour around their nest. When we migrate hives to a new location, we observe that bees are confused and take a day or two to orient themselves to fully orient themselves to their new location.
It is therefore helpful to look at the prospective bee forager. She wanders the dark world of the hive interior for three weeks, that is from the day she emerged from her cocoon. Then, having never before seen the light of day, she takes orientation flights circling and keeping a close eye, much as a feathered fledgling does, on her nest site (Lehrer and Collett 1994). On much closer inspection we discover that bees orient themselves to their front door (Figure 1), flying more or less backwards to teach themselves safe return to their closeted keep. And in so doing, each bee gains confidence in the process learning the defining characteristics of her nest entrance and near surrounds. This she dose by employing a mix of visual and hive odour signals.
Figure 1 Staged turn-back-and-look honey bee orientation learning:
(a) initial bee bee orientation venturing from the nest;
(b) still learning; and
(c) beating a now familiar pathway to and from the nest.
All diagrams: Horridge (209b).
The honey bee turns-back-and-looks instead of relying, as we are so inclined, to glance-back-to-see where we have come from. Bees, crabs, cockroaches and crickets ever only venture out from the shadows once they have a very good notion of safe retreat. To do otherwise would be suicidal. Then, having learnt to recognise her nest, the forager no longer looks back. She, and a menagerie of like-minded sisters, fly directly in as well straight out of their now familiar abode. Now no longer being naïve, these bees have learnt how to stay on track.
But move the hive more than a few meters or reduce the size and shape of their nest entrance – obvious enough to us – and they no longer recognise their abode. They turn to other cues such as hive odour, relearn their surrounds and readjust.
Drifting bees
There is another behaviour, familiar to any seasoned observer, that makes us wonder just how well honey bees recognise features of the complex world they must navigate. This behaviour is tied up in the phenomenon we call bee drift.
Worker bees, and more particularly drone bees, in a crowded apiary are particularly prone to end up in hives other than their own. From this observation and from our knowledge that bees typically forage up to five kilometres from their nesting site, an astounding sixty square kilometre of home territory, we conclude that the ability of a bee to navigate is very good but nevertheless imperfect.
Anatomical cues
That some things must be radically different about insect vision comes further from a simple examination of the anatomy of the bee eye. Add to this the relatively primitive neural setup and one might then concur with Adrian’s observation:
There is no sign that they [bees] reconstruct patterns or put together features to form objects. Bees detect motion but have no perception of what it is that moves, and certainly they do not recognize things by their shapes. Yet they clearly see well enough to fly and find food with a minute brain.
Backhaus (1991) outlines a contemporary understanding of colour vision in insects. But let’s take a close look at compound bee eyes and their photoreceptors. Each worker bee eye has 6900 facets (tiny individual lens), queens about 4500 facets while drones, with more acute vision, 8600 facets. Each of these, comprised of bundles of light tubes or ommatidia, has three distinct light spectrum detectors, one in the blue (λmax 544 nm), six in the green (λmax 436 nm) and one in the ultra violet (λmax 344 nm). Horridge (2016) notes pointedly:
…that attraction of bees to light was controlled by light intensity irrespective of colour, and [that] a few critical entomologists inferred that vision of bees foraging on flowers was unlike human colour vision… and [as noted by Carl von Hess] that bees do not distinguish colours in the way that humans enjoy.
Karl von Frisch, who had so brilliantly discovered the signalling waggle dance, disputed the findings of von Hess, and set out to demonstrate that bees indeed had trichromatic colour vision and see much as we do, a theory that has failed the test of time. Just before the turn of the millennium, it was discovered that bees had a previously unknown type of colour vision, more correctly several types of vision, extraordinary findings that deserve some explanation.
Blue as a visual recognition tool
Firstly, bees discriminate between different levels of blue saturation (or tone of blue) and indeed can recognise blue objects. White surfaces have a high blue component, so bees can easily recognise a hive painted white as their home but have no ability to discriminate between similar hives based on this ‘blue signal’ alone.
Bees returning to a large paddock apiary observe a ‘sea of blue’ so need other landscape cues to guide them safely home. White beehives scattered in woodland will more likely appear as a series distinctive beacons to the weary returning bee.
But can changing the saturation of blue by painting hives different shades of blue allow bees to clearly discriminate one hive from another in the types of crowded setup shown in Figure 2? In this situation we should note that recognition of blue is not so much as ‘seeing objects as blue’ as it is about integrating the total amount of blue in objects in a limited, but defined, field of vision.
This suggests that one cannot assume that bees will discriminate individual hives as ‘separate objects’. More widely spaced hives might be more readily learnt to be ‘separate targets’ with ‘different total blue signals’.
The take home message is that bees can recognise the amount of blue in an object and to some extent a height component but that this may be more of a beacon than a hive identifying tool.
Figure 2 Does ‘total amount of blue’ help bees to recognise their hive under crowded conditions? Note hives painted white appear deep blue to bees.
Images: Alan Wade.
Green and blue edging effects as a navigation tool
Secondly, bees employ their green, and to a much lesser extent blue, receptors to recognise edges and shapes. Flying bees scan edges of dark objects (poles, hedges, tree canopies...) that generate a green interference fringe. They ignore and cannot see green objects such as trees or grass, instead integrating edges of dark objects that appear momentarily as a green fringing light chimera. Remarkably bees have no memory or capacity to recognise a green as a colour. Nevertheless bees detect blue and green light. In a parting meeting Adrian told us that bees are dichromatic, a statement we accept as an article of faith: 1+1=2.
The take home message is that bees can detect edges, their length, orientation and shape, in flight. They integrate these ‘edge signals’ to recognise flowers and other objects, such as a hive entrance, also using this ‘green signalling’ to navigate safely over long distances.
Ultraviolet as a directional navigation tool
Thirdly bees make use of polarised light to orient flight. This sensing works on ultraviolet sensing and employs sky light, that is directionally filtered and invisible to us, as a directional compass. They do not process this signal in the sense that blue and green light are relayed back to the brain.
The take home message is that bees use polarised light for directional signposting.
Vertebrate versus invertebrate sight
So what to make of this revolutionary notion of honeybee sight, one shared in various measure by most sighted invertebrates? We have found it helpful to note that much of our response to visual stimuli is often grounded less by the images we see and more in the what-have-we-got-here reflex. We see much from the corner of our eye: the child on the children’s crossing, the rat disappearing down the drain, and the diving magpie, useful enough cues for action. Often enough we react to these cues without having any particular notion of what we might or might not have seen. We also see black (an absence of light stimulus) and white (not a colour but a construct of response to a multitude of colours) that evoke a very different response in bees.
We can summarise the differences between what we and many vertebrates see (in large measure images) and what invertebrates such as bees see (a simple integration of specific light signals) in terms of the light photons received:
Blue Bees see blue not so much as a patch of blue but as a summation of all the blue content in their line of vision. Further bees see the blue component of many objects we perceive to perceive as another colour shade. For example white light has a very high blue component and shines vividly blue to bees, something we gain an inkling of from our observation of the prism effect of air generating rainbows and in air movement in creating mirages.
Red and yellow While we see the primary colours red and yellow, bees have no photoreceptors for this part of the light spectrum, so are totally invisible to bees.
Black Black is a construct colour (absence of light) to us but does not register with bees.
Green Bees detect and sum contrasting green fringes on the edge of dark objects generated in flight but are not sensitive to green objects. They use their six green detectors, each with imbedded directional elements, to navigate but cannot see green as we do.
Ultraviolet Like green, bees employ ultraviolet light to navigate but in a different way. Shifts in polarised sky light, demonstrated by other invertebrates response to bright starlight, are a directional signpost. While being unable to see ultraviolet light, leaving us, so to speak, in the dark, we sense the damage that can be inflicted by ultraviolet light and use sunburn cream and avoid looking directly at the sun.
Infrared Bees, like us, have no infrared photoreceptors. Humans are nevertheless extremely sensitive to hot and cold surfaces detecting minute changes in temperature with ease. Beekeepers have taken to using infrared cameras to detect heat patterns and serious heat losses from bee colonies trying to overwinter (Figure 3).
Figure 3 Late winter beehive imaging through:
(a) the lens of an infrared camera; and
(b) the lens of a visible spectrum camera.
Note the infrared camera output depicts warm or hot as red, grading though our visible spectrum to yellow as cooler and green near ambient and blue as cold.
Images: Alan Wade.s
Bees have a remarkable facility to closely regulate brood and swarm cluster temperatures so must have an ability to detect the infrared if not visually.
The honey bee visual system discovered
We have outlined bee use of blue, green and ultraviolet light receptors to navigate. But how were these discoveries made?
It has long been known that bees can be trained to choose between two targets, only one of which is rewarded by sugar syrup. Adrian designed many such targets, based on his knowledge of invertebrate colour photoreceptors and the known colour makeup of standard colour papers, to train bees to discriminate between different patterns such as simple black bar stripes painted on a target.
Armed with the knowledge of insect physiology and a detailed understanding of the structure and neural connections of eyes in invertebrates, Adrian was able to focus on the capacity of bees to recognise form and shape of objects. He was unencumbered by the many confounding theories of insect colour vision that had beset the majority of earlier bee vision investigations.
Training bees
In an elegant experiment actively foraging bees were first trained so that they could learn to recognise, and avoid, a blue target (Figure 4). Bees can’t see black and bees do not see yellow, so the training exercise was exceptionally simple.
Figure 4 Training foraging bees to ignore the green fringing effect of solid black edges and to avoid unrewarding blue:
(a) target setups as we see them; and
(b) target setups based on reflected light signals available to bees.
Horridge had learnt that bees used green receptors to detect edges of dark objects (despite the fact that black is invisible to bees). However in this setup bees learnt to ignore green interference fringes as both had identical black outlines. Figure 4a redrawn as Figure 4b represents the light patterns scanned by foragers.
By not only disabling the green receptor mechanism for making an effective choice, bees also learnt to avoid the blue to successfully harvest sugar syrup.
Testing time
So trained, the same bees were now asked to sit ‘flying exams’ to discover what they had learnt.
In the first test (Figure 5), the blue colour signal is entirely removed so the bees could no longer rely on avoiding blue to be rewarded: Figure 5b shows that blue signalling is entirely blocked out and also shows how green signalling is removed from the unrewarded target.
The bees, having learnt to completely ignore green, and no longer being able to rely on avoiding blue to be rewarded, now mainly avoided the reward target as it generated all the green signal.
To restate this outcome, foragers mainly visited the grey target, a form of diminished white with truly minimal blue content and little or no capacity to generate the green contrasting fringe effect.
Figure 5 Testing what bees have learnt by removing blue signal:
(a) setup where blue square is replaced with a grey slate; and
(b) same setup redrawn to depict what signals bees receive.
We have observed bees foraging patches of flowers in open parkland. They stay momentarily on some blossoms but drink deeply or busily comb anthers on other flowers, suggesting some measure of randomness in targeting floral rewards. This may account for the 30% of avoidance of learnt cuing or it may reflect differences in the capacity of individual bees to reliably learn patterns.
But what happens if, instead, the invisible yellow panel were replaced with a blank white square (Figure 6). The green fringing effects of solid black bars are unchanged so the trained bees can simply ignore green signalling in making a choice between targets. But the use of a white – to replace yellow – panel results, as we have learnt, in bees seeing white as a vibrant blue (Figure 6b).
Surprisingly to us, but not to the bees, the majority of bees now visit the relatively pale blue unrewarded target.
Figure 6 Testing how bees respond to changing the amount of blue in a target:
(a) the rewarded yellow panel is replaced with a white panel; and
(b) same setup redrawn to depict what signals bees receive.
In a final test bees were challenged by more ambiguous targets (Figure 7). Blue is entirely removed from the equation so foragers can no longer rely on avoiding blue to choose the rewarding target so default to reliance on green fringe, a cue they observe but were trained to ignore.
Figure 7 Testing how bees respond when blue signalling is switched off:
(a) setup where the blue square is replaced with a black slate; and
(b) same setup redrawn to depict what signals bees receive.
The green fringing effect is reduced, but not removed, so that foragers are confused, randomly visiting both targets (Figure 7b).
So we confirm the capacity of bees to learn very specific signalling, probably most useful in targeting floral resources and in making the final dash to carry their treasured cargo to their nest entrance. In a wide range of experiments, Horridge demonstrated that bees could discriminate a wide variety of other patterns, e.g. small and larger star shapes and circles, and could integrate broken edges and edges with different orientations.
However these patterns, though clearly learnt, belie the complexity of the natural environment so simple extrapolation of experimental findings to the real world signals caution. By way of example, bees returning to their nest can be trained to recognise a circular hive entrance. They easily recognise a dark circular perimeter (using their green fringing facility) and will not hesitate to enter the hive even if the diameter of circular entrance is increased substantially. However they will baulk, but can relearn to enter the hive, if the entrance is reduced substantially. This finding is not dissimilar to our every day experience of moving hives just a few meters daily and watching – obvious to us – their changing circumstance.
These, and a multitude of other trials, showed that bees sum green interference edge lengths of landscape features, the blue content and height of objects, and the signpost of sky ultraviolet light to cue their whereabouts.
With very limited neural capacity and a known tendency to sum up the light stimulation signals, bees ignore information that is not useful to them, thus avoiding overload.
Practical applications of the honey bee vision system
Apart from having gained a basic understanding of the complexity and nature of the visual world of the honey bee, we would be remiss if we were we not to ask how some of this ‘new knowledge’ might be used to improve our stewardship of honey bees.
In the truck spotlight
In our quest to comprehend bee vision, we were asked by an old queen breeder pal Frank Malfroy (pers, comm.) how he might turn off his headlights – well to bees – when moving his bees at night:
We’ve got lights on our machine and truck which the bees are attracted to. If I could put a coloured lens on the machine lights it would make it much more comfortable for the operator...
Here the answer is seemingly simple. Employ strong yellow lights (red lights might be mistaken for taillights) making sure there is no hint of stray white or blue light. In practice however, the light that appears intensely yellow to us may contain other spectral elements, those that invertebrates like moths and bees can detect at low levels. So we cannot assume that bees will not be attracted to a standard yellow light. Adrian is no longer with us to pose these inferences, but beekeepers extracting honey at dusk well know not to switch on lights until it gets dark and bees have returned to their nest.
Marking time
John Free (1958) conducted experiments to characterise inter-colony honey bee movement, bee drift, concluding that:
most of the bees which drift do so... before they become regular foragers;
bees emerging in August and September [our winter-early spring] drift less than those emerging earlier in the year [and presumably because they are long familiar with their hive];
drifting varies considerably in different circumstances, and may be extensive;
drones drift two to three times as frequently as workers [likely because they so readily detect queen pheromones];
an individual bee is more likely to drift from a small to a large colony than vice versa, but the greater number of bees flying from the large colonies may result in a net gain in bees to the smaller hives;
when hives are arranged in repetitive patterns, bees drift to hives occupying similar positions in the pattern to their own. When hives are arranged in rows, bees from the centre colonies drift more than those at the end, resulting in the latter colonies gaining numerically. In some circumstances, more bees drift to hives in one direction than in the opposite direction; and
facing hives in different directions and painting them different colours [we now note that bees have very limited colour discrimination capability] considerably reduces drifting, the facing of hives in different directions being of the greater significance.
So might we use the green signal, by applying back tape to our bee boxes, or paint them various hues of blue to improve the homing instinct of returning forager bees?
It seems unlikely that the problem of drift in a crowded apiary, one that Tom Seeley has signalled is inherent to any mass aggregation of hives, is resolvable. Bees returning to a crowded apiary seeing a sea of white hives (to the bee a large number of blue beacons) poorly defined by landscape features will not discriminate between one hive and the next. Furthermore distinctive marking patterns would likely not assist bees when they faced a multiple choice between a large number of hives.
Nevertheless we asked ourselves whether distinctive marking patterns might assist the homing bee where there are a few, probably less than three of four, hives sited on a single apiary pad. Bees are well able to learn to discriminate between one and two vertical bars (by summing the lengths of green interference fringe, Figure 8a) and learn to differentiate a left and right boundary marker to the left or right (Figure 8b) (Horridge, 2015) and, as we have noted, can recognise a variety of shapes (e.g stars, circles and sloping lines).
Painting bee boxes the colours of the rainbow and marking them in a random manner will do nothing to reduce bee drift or mated queen loss (Horridge, 2017). Seeley (2019) noted widely spaced wild nests in a woodland provide a rich cuing system, one that protects them from robbing, bee drift and, in some measure, ready disease transmission.
Figure 8 Black stripe marking schemes (e = hive entrance) to enable bees to better integrate green fringe and distinguish between hives using:
(a) summed edge lengths; and
(b) left-right polarity edge discrimination.
We think that a simple marking scheme to reduce drift may work in a hive splitting exercises or in a back yard operation where just a few hives are crammed together (Figure 9) but we do not say this with any certainty.
Much can be learnt from queen breeders reliant on having newly mated queens getting back to their hives reliably, that is, not drifting to the wrong hive leaving their hive queenless. They put out their nucs with plenty of landmarks spacing them widely enough to limit drift and return of newly mated queens to the wrong hive.
Despite the difficulties of organising large numbers of hives in paddocks, it would appear that abandoning the practice of lining up hives in rows and where possible spreading them out more could reduce the incidence of disease transmission and increase productivity. Cale, Banker and Powers (1975) describe the practical consequences of drift, those primarily associated with crowded apiaries, confirming John Free’s experimental findings.
Figure 9 Small hive group with discernible marking system but uncertain hive spacing requirements.
Shaping up for size
Its helpful to understand bee edge detection facility in contexts other than bee nest recognition. Bees, given a choice, won’t fly though a dense thicket: they will fly over vegetation, over a fence, or around chicken wire barrier. Netting should ever only be placed over fruit trees until after fruit set.
In seemingly complex ways bees employ their left-right and up-down green sensors to measure shape and edge lengths to navigate around trees, hedges and large objects, to recognise and target individual rewarding flowers and to cue nest location. Such cuing may be augmented by factors such as hive odour and fragrance elements of flower exudates.
Our journey to gain some understanding of bee vision has led us to some surprising reevaluation of the role bee sight plays in their interaction with their environment. Bees cannot see other bees, beekeepers or the meadows they so purposefully forage. They visually cue their environment in ways that are distinctly unfamiliar to us but in ways that suit their needs to forage and propagate. But you can never tell with bees.
Just on the offchance you might doubt the veracity of Adrian’s insights we heard tales of crickets losing their visual amenity after being fed ad lib in the dark after several generations while crabs and dung beetles could read the sun, the moon and the stars. For a lucid description of the crab Carcinus maenas ability to see celestial light dial up Adrian’s published study of six decades ago (Horridge, 1965, see references for download link).
Sometimes rewarded by Adrian’s explanations, we came away from our mid morning visits not so sure we had full grasped the nettle, a feeling borne out in the 1973 Schlock movie:
I feel more like I did when I came here than I do now.
Acknowledgement
We would like to thank Emeritus Professor Horridge for extensive discussions and critical reviews of our manuscript.
Appendix (Brief summary of bee vision supplied by Adrian)
A summary of honeybee colour vision
(As published in recent papers and the CABI book; 2019).
In the study of the bees’ vision of colour, first the emission spectrum of each of the standard coloured commercial papers in local sunlight was measured. The result for each paper was multiplied at each wavelength by the spectral sensitivity of each of the three types of receptor cell, which had been measured with a microelectrode. Each ommatidium usually contains one receptor cell optimally sensitive to ultraviolet (UV) one to blue, and six to green light. As shown first by Mathilde Hertz (J. Exp. Biol. 16, 1-8. 1939; and Naturwissenschaften 25, 492. 1939), and many others although they were unaware of it, UV inhibits the detection of white and therefore of blue, so that trichromatic colour vision is impossible. The receptor with peak in the blue detects areas, where the blue content is summed, and the average height of blue can be learned. Bee vision of areas is monochromatic in blue, but each colour has a different content of blue, as compared to the background of green.
Six receptor cells with maximum sensitivity to green light feed into neurons of the lamina that detect green contrast only, excluding the tonic part of the signal. The bee is able to learn green modulation which is equal to the total of (length of edge multiplied by the contrast at each piece of the edge) so that pattern is lost. Similarly, blue contrast at edges is summed to give a blue modulation signal. Green modulation inhibits blue modulation; therefore, vision of edges is usually monochromatic in green. There are no receptors for black or white, and no achromatic vision of black or white patterns.
All signals at this level have the spatial resolution of the retinal array, but in each case, the signals of each kind are summed over large areas of the eye to maximize sensitivity in certain responses. This summation has no effect on detection of a striped grating, but interferes with pattern discrimination.
Bees detect the horizontal position of green modulation, and are able to use this to measure width of a vertical bar, the separation betweentwo bars, or the asymmetry of green modulation in the horizontal plane.
Edge detectors are very short, only three ommatidia wide on the retina, and are not strung together to signal long edges. Instead, thebee takes an average of the orientation within quite large areas, so that equal lengths of edge at right angles cancel to give zero orientation.
Bees also detect a circle of edges, and can learn the position of the centre.
Bees readily detect coincidences between the responses of feature detectors, especially the polarity of blue relative to a vertical green edge or a landmark of green modulation. Cues with polarity indicate a direction or a turning point, like a signpost, as opposed to a symmetrical cue, which is more likely to indicate a destination, like a shop-sign.
Readings
Backhaus, W. (1991). Colour vision in insects. In Gouras, P. editor. Vision and visual disfunction. London: Macmillan. pp.262–288. in Hempel de Ibarra, N., Vorobyev, M. and Menzel, R. (2014). Mechanisms, functions and ecology of colour vision in the honeybee. Journal of Comparative Physiology. A, Neuroethology, Sensory, Neural, and Behavioral Physiology 200(6):411-433. https://doi.org/10.1007/s00359-014-0915-1
Cale Sr, G.H., Banker, R. and Powers, J. (1975). The Hive and the Honey Bee, Revised Edition, 358pp. Chapter 12, Management for honey production: Swarm prevention and control, pp.380-384. Dadant & Sons, Hamilton, Illinois.
Free, J.B. (1958). The drifting of honey-bees. The Journal of Agricultural Science 51(3):294-306. https://doi.org/10.1017/S0021859600035103 https://www.cambridge.org/core/journals/journal-of-agricultural-science/article/drifting-of-honeybees/88DA09239D5DC14BFA09A6F2BD99DF2F
Horridge, G.A. (1965). Direct response of the crab Carcinus to the movement of the sun. Journal of Experimental Biology 44(2):275-283. https://sci-hub.usualwant.com/10.1242/jeb.44.2.275
Horridge, A. (2015). How bees discriminate a pattern of two colors from its mirror image (2015). PloS ONE 10:1-23. https://doi.org/10.1371/journal.pone.0116224
Horridge, A. (2016). Parallel inputs to memory in bee colour vision. Acta Biologica Hungarica 67(1):1-26. https://doi.org/10.1556/018.67.2016.1.1 https://pubmed.ncbi.nlm.nih.gov/26960353/refers Hess 1917 Menzel R. and
Backhaus, W. (1991). Colour vision in insects. In Gouras, P. editor. Vision and visual disfunction. London: Macmillan. pp.262–288. in Hempel de Ibarra, N., Vorobyev, M. and Menzel, R. (2014). Mechanisms, functions and ecology of colour vision in the honeybee. Journal of Comparative Physiology. A, Neuroethology, Sensory, Neural, and Behavioral Physiology 200(6):411–433. https://doi.org/10.1007/s00359-014-0915-1 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4035557/#:~:text=The%20honeybee%20eye%20contains%20three,2
Horridge, A. (2017). Why newly mated queens get lost. American Bee Journal 157(9):985-986. http://adrian-horridge.org/downloads/Recent%20papers%20on%20bees/LostQueensAmerBee%20J.pdf also published as Horridge, A. (2017). Why newly mated queens get lost. The Australasian Beekeeper 118(12):544-545.
Horridge, A. (2018). Bee vision is totally different. Lecture to Canberra Region Beekeepers annual general meeting, 19 April 2018.
Horridge, A. (2019a). A summary of honeybee colour vision. http://adrian-horridge.org/downloads/A%20summary%20of%20honeybee%20colour%20vision.pdf
Horridge, A. (2019b). The discovery of a visual system: The honeybee, 256pp. CABI, Wallingford, Oxfordshire.
Lehrer, M. and Collett, T.S. (1994). Approaching and departing bees learn different cues to the distance of a landmark. Journal of Comparative Physiology A 175:171–177. https://link.springer.com/article/10.1007/BF00215113
Milne, A.A. (1956). Winnie-the-Pooh. Chapter 1. In which we are introduced to Winnie-the-Pooh and some bees, and the stories begin. Methuen & Co. Ltd., London, 36 Essex Street, W.C.2http://www.longwood.edu/staff/mcgeecw/Poohchapter.htm
Seeley, T.D. (2019a). The lives of bees. The untold story of bees in the wild. Princeton University Press, Chapter 16, Colony defense, pp.260-263.
Comments