Scent of a man

Al Pacino has received five nominations in the Academy Awards 'Best Actor' category. Although arguably not his best film, his only win - at least, so far - was for his role as Lt. Col. Frank Slade in the 1992 film, Scent of a woman {{1}}.

Slade plays a decorated, blind, alcoholic and suicidal, retired Vietnam veteran. It might not sound appealing, but it is a good film and contains several well-known scenes. In one, he notices the perfume of a woman in a restaurant - hence the title - and offers to teach her to tango.

Being blind, Frank is even more dependent upon his other senses to detect changes in his environment.

Honey bees, which spend the majority of their lives in the dark confines of the colony are - at least in some ways - similar, though they are not blind {{2}}. Although they can visually discriminate between colours and patterns, they cannot distinguish between bees from different colonies by sight. Instead, they rely upon the 'smell' of the cuticular hydrocarbons (CHC) that are characteristic and, at times, unique.

Cuticular hydrocarbons

I've written about CHC's and workers before. It's an interesting topic; CHC's are impacted by the microbial population of the gut, and - in turn - influence both the ability of young flying bees to drift between hives, and the recognition of non-nest mates by guard bees. {{3}}.

None of which tells you what CHC's actually are.

This is how I described them in a previous post:

CHC’s are long-chain hydrocarbons (predominantly alkanes, alkenes, and branched alkanes) made in [special cells and] transported to the outer cuticle of the insect where they form a water-repellent layer that helps prevent desiccation.

CHC’s are also pheromones and the ratios of the various hydrocarbons in the mix differ between colonies, providing a unique ‘fingerprint’ that allows guard bees to discriminate between nest mates and non-nest mates.

The names (alkanes, alkenes etc.) don't matter ... just think of them as subtly different chemicals, which - where needed - I'll distinguish between using a number. For example, C23 and C24 are both CHC's, but differ in the number of carbon (C) atoms in the chain (one has 23 and the other - spoiler alert - has 24).

Workers are, of course, female ... but the title of this post is 'Scent of a man'.

That's because today I'm going to write about drone CHC's, specifically with reference to some drone behaviour we'll see a lot of in the next few weeks, and some drone behaviour we've (or at least I've) seen too little of over the summer.

I'm referring to the ejection of drones in the autumn, and queen mating in the summer, respectively.

It's a drone's life

Drones are produced in the colony from mid-Spring until late-Summer {{4}}. I've discussed the critical timing of drone production in another post, so won't rehash things here.

After emergence, a drone takes 9-12 days to become sexually mature. Once mature the drone embarks on daily flights in search of a virgin queen to mate with. These flights are not random. They generally occur in early afternoon and involve the drone visiting one (or often more) drone congregation areas (DCA's). These are geographically-defined locations, conserved year to year, which the virgin queens also visit and within which mating almost always occurs.

A single DCA might contain thousands of drones and - on a calm, warm, summer afternoon - you can hear them if you are nearby; they sound like a small swarm as the drones circle around, 5-40 metres up, waiting for a virgin queen to chase.

Most drones perish on these flights - they either die trying to mate, or die when mating.

Get out and don't come back!

However, during times of nectar dearth, or towards the end of the season, drones are an unwanted drain on the resources of the colony. In particular, older drones are surplus to requirement, and the workers in the colony eject them. Without food, they quickly perish.

Drones are being ejected from my hives now {{5}} ... colonies are contracting in size, the nectar flow is minimal, the days are shortening, and the weather is even less conducive to queen mating than it was in mid-Summer.

Workers either prevent returning drones from re-entering the hive, or they forcefully harass and drag drones from inside the hive out of the entrance.

One final point to make about drones; their motto is "Wherever I lay my hat (That's my home)". They drift (apparently) freely between colonies. If the recipient colony is not actively ejecting drones it appears as though drones from other colonies are welcomed, or at least accepted without much scrutiny.

How can these behaviours be explained?

Well, inevitably, as these behaviours are complex, the explanation is also likely to be complex or multifactorial. However, there are some hints that drone CHC's could, or may, influence ejection from the colony, attraction to DCA's, and attraction of queens within the DCA {{6}}.

And, in the same way that Scent of a woman dates back 2-3 decades, so does the beginning of this story, with the analysis of drone CHC's by Wakonigg and colleagues in Karl Crailsheim's laboratory (Wakonigg et al., 2000).

Detecting and identifying CHC's

My just-barely-passed GCE 'O-level' in chemistry makes me laughably unqualified to discuss this is any detail. Fortunately, that's not a problem; the gory details are unimportant to understanding the overall story.

Suffice to say, CHC profiles (i.e. the range and quantity of each) can be determined using a method called gas chromatography, and the individual constituents unambiguously identified using mass spectometry. {{7}}

Wakonigg et al., (2000) allowed drones from three unrelated colonies to emerge in cages in the incubator, marked the thorax with paint to indicate their day of emergence, and then returned them to the hive they were reared in until needed. They subsequently analysed the CHC's in drones of a range of ages, though I'll restrict most discussion to pooled groups of 1-3 ('hive drones'), 6-10 and 12-19 days old ('flying drones'). They also analysed just-emerged drones (0 days old).

Unlike workers and queens, drones emerge and, after 3-4 days go on a series of orientation flights. They then remain in the hive until they reach sexual maturity, 9-16 days after emergence. Therefore, the four groups tested (0, 1-3, 6-10 and 12-19) represent drones that were unexposed to the hive (0), immature (1-3), flying and maturing (6-10) and sexually mature (12-19), respectively.

Similar, but different

The authors clearly don't believe the idiom that "a picture is worth a thousand words" as the results are presented in a series of turgidly monotonous tables.

However, the results can be summarised as showing that:

1. CHC profiles of drones of the same age from different hives were broadly similar, and
2. CHC profiles change as drones get older.

There, that wasn't so difficult, was it?

But, since I firmly believe that "a picture is worth a thousand words" (particularly if I write them), here are two graphs of mine and the one figure from the paper that nicely illustrate these two points.

In the graphs above I've plotted the quantity of different CHC's in drones of 0 days old (left) or 12-19 days old (right) from the three different hives (coloured blue, green and red). Although there are differences, drones of the same age usually have similar quantities of the individual CHC's.

However, when drones of different ages are compared, specific CHC levels may differ; e.g. compare the levels of C25 in 0 and 12-19 day old drones.

Interesting ...

Because of this similarity between CHC profiles in drones of the same age from different hives, and the change in CHC profile as the drones age, it is possible to predict the age of a drone from it's CHC profile.

The CHC profiles of 223 drones of known age were used to predict the age of 50 additional drones (the age of which were also known, enabling them to validate the prediction). The graph above shows these 50 drones (as points) with the line indicating the linear model of the changing CHC profile. The age could be predicted with a standard error of ~2.8 days.

Drones from different colonies have similar CHC profiles which change predictably as the drones age.

CHC profiles - the scent of a man drone - are therefore predictive of the age of the drone.

Could CHC's explain drone behaviour and ejection?

The similarity in CHC profile of drones of a similar age suggests to me that workers - whether guards or nurse bees - would be unable to discriminate between drones from different colonies.

Perhaps this explains how easily drones drift between colonies?

Although I can see evolutionary advantages to a colony that only allowed access to drones from the same colony, or fed those drones once they had accessed the colony {{8}}, perhaps these are outweighed by the need for multiple, genetically-distinct, matings for colony fitness?

Alternatively of course, CHC's may have no role in acceptance or rejection by guard or nurse bees. Instead, there might be some sort of 'switch' that operates ... "It's a drone, ignore the smell".

What about ejection of old drones from the hive?

Old drones are of little use to a colony. About four weeks after emergence they are too old to mate, although they can live for at least 55 days (and possibly more in queenless colonies or overwinter).

It seems possible to me that the different CHC profile of elderly drones (it's a pity Wakonigg et al., didn't analyse drones of more than 19 days of age) may trigger rejection by workers from the colony towards the end of the season, or potentially during periods of nectar shortage.

This would be relatively easy to test, though I'm not aware that the study has been done. The similarity of the CHC profile of drones from different hives would facilitate exclusion of the decrepit ones irrespective of where they were originally reared, whilst potentially 'saving' the younger ones to mate another day, once conditions improve.

Behavioural differences

Most of the section above is speculation. It's worth noting that there are behavioural differences in drones of different ages which might allow the workers to determine whether they should be ejected or not.

For example, drones are temperature sensitive (Crailsheim et al., 1999). Younger drones prefer warmer temperatures, so are attracted to the centre of the brood nest (where they can be fed and supported). In contrast, older drones migrate to the periphery.

I'm sure you've opened hives to find the outer frames - well away from the brood nest - covered in adult drones. These are the older, sexually mature, drones (and presumably knackered old, pensioner drones, too old to mate). Drone eviction by workers might simply involve 'picking on' any drone outside the core of the colony ... it doesn't have to involve CHCs at all.

However, if there is any age-restricted prevention of access of drones to the colony, this obviously cannot be determined by their geographic location within the colony.

More research is needed.

Drone congregation areas

Drones live for weeks, and very rarely overwinter. However, drone congregation areas (DCA's) exist for decades or longer (there's one described by Gilbert White's Natural History of Selborne published in 1789 that still exists).

How do drones 'find' DCA's? You can cause drones to congregate using queen pheromones, but DCA's are occupied by drones before the virgin queens arrive to mate.

DCA's therefore exist in the absence of queens, and queens also need to 'find' them.

Several studies have implicated visual clues on the horizon (mountains, valleys, forests) or terrestrial anomalies in magnetic fields, as 'defining' DCA's, but their locations are difficult to predict and these relatively crude geographic features cannot explain the large, very localised, concentrations of drones observed. A single DCA may contain 11,000 drones from well over 200 colonies {{9}}.

Unsurprisingly, as they're involved in so many other aspects of the biology of honey bees, there has long been a suspicion that the formation of DCA's may involve pheromones.

CHC's are pheromones.

However, although DCA's are localised and might contain thousands of drones, they are still relatively difficult to locate, large - imagine a cone 10-40 metres in height, or 100-200 metres diameter - and transient so it is tricky to sample and measure pheromone concentrations within them.

Instead, scientists have developed systems to quantify odours that are attractive to bees using a so-called walking simulator.

A bee is tethered above an air-supported ball that she (or he) can walk on, so turning the ball. The direction of walking is recorded by computer in a manner similar to the way a computer mouse moves the cursor on screen. The bee is exposed to odours in the airstream that passes over the ball. The bee walks towards attractive odours, or away from unattractive ones. Experiments are conducted in the dark.

Drones attract drones and queens

I'm going to summarise a lot of research in a few short paragraphs here as all the studies use the walking simulator described above (and were conducted in the same laboratory, that of Jean-Christophe Sandoz at CNRS in France).

1. Brandstaetter et al., (2014) demonstrated that drones are attracted to 9-oxo-2-decenoic acid (9-ODA), the queen-produced sexual attractant pheromone, but not to queen mandibular pheromone (which dominates the odour of mated queens). The attraction to 9-ODA was unsurprising; it was known that drones use this pheromone together with short-range visual clues to locate the queen within the DCA. However, Brandstaetter et al., (2014) also showed that drones were attracted to other drones, but not to the odour of workers, in the walking simulator assay. They suggest that drone-emitted pheromones - which include CHC's - might be involved in DCA formation.
2. Bastin and colleagues (Bastin et al., 2017a) demonstrated that virgin queens were attracted to the odour of drones, but not workers, in walking simulator experiments.
3. In an extension of the 2014 study, Bastin et al., (2017b) demonstrated that the attraction of drones to other drones was an age-dependent phenomenon. As part of this analysis they also confirmed that the odour profile - partly defined in terms of CHC's and potentially involving other drone-produced pheromones as well - changed as drones matured (so confirming the earlier results from Wakonigg et al., 2000). The attraction of drones to other drones was dependent upon the drones being sexually mature.

In the context of the mating biology of honey bees these three results are very interesting and suggest a potential role for pheromones, including CHC's, in the aggregation of sexually mature drones within the DCA, and the attraction of virgin queens to pre-formed DCA's.

Questions, questions, questions ...

As with all research - or at least all interesting research - a single result prompts dozens of follow-up questions.

The more you know, the more you realise you don't know {{10}}.

The walking simulator assays suggests that there are drone-produced odours that attract other sexually mature drones, and virgin queens.

• Are these attractant odours CHC's or are they other drone-produced pheromones, such as drone mandibular pheromone?
• What are the active ingredients within the odour that attracts other sexually-mature drones, and are they the same active ingredients that attract virgin queens? For example, older drones produce higher levels of C25 and C33:1 ... is just one of these sufficient, or a cocktail?
• And before we answer that, are virgin queens only attracted to sexually-mature drones? Perhaps they just like the 'Scent of a man'.
• As described above, drones also congregate within the hive. Do volatile odours like CHC's contribute to this type of congregation, or is it solely a consequence of the reduced peripheral temperature outside the brood nest?
• Queen mating is associated with warm, dry, calm weather. Is mating less successful in windy conditions because it results in the spread or dilution of drone pheromones?

I could go on, but won't.

Drone congregation areas warrant an entire series of posts of their own, particularly since mechanical flying drones can now be used to more-easily locate them {{11}}.

Check your hives over the next few weeks. Watch the entrance activity and observe those doomed drones being evicted ... potentially because they no longer have the scent of a man.

Sponsor The Apiarist

The Apiarist covers 'the science, art and practice of sustainable beekeeping ... so much more than honey' and is only readable because of the many hours spent frantically correcting hte tpyoes generated by my fat fingers.

If this, or other, posts has helped or inspired your beekeeping, or was amusing, informative or enlightening, then why not become a paying sponsor of The Apiarist?

Sponsorship costs less than £1/week annually, or a little more than a large moccha monthly. Sponsors receive the weekly posts, an irregular monthly newsletter, and an increasing number of sponsor-only content ... those starred ⭐ in the lists of posts.

Alternatively, help reduce my outrageous caffeine overdraft ... and please spread the word to encourage other beekeepers to subscribe.

Thank you.

References

Bastin, F., Cholé, H., Lafon, G., and Sandoz, J.-C. (2017a) Virgin queen attraction toward males in honey bees. Sci Rep 7: 6293 https://www.nature.com/articles/s41598-017-06241-9.

Bastin, F., Savarit, F., Lafon, G., and Sandoz, J.-C. (2017b) Age-specific olfactory attraction between Western honey bee drones (Apis mellifera) and its chemical basis. PLOS ONE 12: e0185949 https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0185949.

Brandstaetter, A.S., Bastin, F., and Sandoz, J.-C. (2014) Honeybee drones are attracted by groups of consexuals in a walking simulator. Journal of Experimental Biology 217: 1278–1285 https://doi.org/10.1242/jeb.094292.

Crailsheim, K., Eggenreich, U., Ressi, R., and Szolderits, M.J. (1999) Temperature Preference of Honeybee Drones (Hymenoptera: Apidae). Entomologia Generalis 37–47 https://content1.schweizerbart.de/papers/entomologia/detail/24/81786/Temperature_Preference_of_Honeybee_Drones_Hymenoptera_Apidae?l=EN.

Wakonigg, G., Eveleigh, L., Arnold, G., and Crailsheim, K. (2000) Cuticular hydrocarbon profiles reveal age-related changes in honey bee drones (Apis mellifera carnica). Journal of Apicultural Research 39: 137–141 https://doi.org/10.1080/00218839.2000.11101033.

{{1}}: In my view, his over-the-top depiction of Tony Montana in Scarface, the doomed and ageing Lefty Ruggiero in Donnie Brasco, or Michael Corleone in The Godfather were probably all better.

{{2}}: Nor are they decorated or alcoholic ... but I am preparing a post on suicide in social insects for sometime this winter.

{{3}}: Remember also that rejection of non-nest mates by guard bees is itself influenced by whether there is a nectar flow or dearth.

{{4}}: And at other times if the queen is a drone-layer, or if there are laying workers in a queenless colony, but all subsequent discussion relates to normal, queenright colonies.

{{5}}: Well, not now as it's 11 pm, but certainly this week.

{{6}}: Could meaning I can envisage a scenario in which they are involved (although there's no current evidence that they are), and may meaning that there's at least some direct evidence supporting their involvement.

{{7}}: That's all you need to know ... you can thank me later 😉.

{{8}}: Until they are ~7 days old drones beg for food from nurse bees and are partially or wholly supported by the colony.

{{9}}: By my calculation, a typical DCA - 35 metres high and with a maximum radius of 75 metres - has a volume of about 200,000 m³, so the concentrated drones aren't actually very concentrated (however, they are not evenly distributed).

{{10}}: Variously attributed to Aristotle, Socrates, Einstein, George Bernard Shaw and - possibly - Homer Simpson.

{{11}}: At least, more-easily than scampering about waving a fishing rod over your head.