Soldier bees, super-soldier ants

by Matthew Cobb

One of the most amazing things about social insects – bees, wasps, ants and termites (hornets are just big wasps) – is the way they have a division of labour. At its simplest, this division of labour is morphological and involves having a reproductive individual or group of individuals who is the queen. She is genetically identical to all the other females in the colony/hive, but she is specialised for producing loads of eggs. We know there’s nothing genetic about being a queen – there’s no special ‘queen gene’. It’s simply a matter of how much food she is given when she’s a larva. The same seems to apply to those ant species that have specialised workers – ‘soldiers’ – that carry out defensive or aggressive tasks. There’s no gene involved in being a soldier, it is environmentally determined.

This raises a fascinating question: how does the genome of a single species code such a wide variety of forms? In other words, how are the different developmental pathways turned on and off?

Things get even odder when you realise that although many ant or termite species have ‘soldiers’, there are no known cases of bees or wasps having morphological worker casts. There are no ‘soldier bees’ or ‘soldier wasps’. Why not?

Two recent papers – one in Science on ants, the other on bees in Proceedings of the National Academy of Sciences (USA) (PNAS as it is generally known) – have addressed these issues, giving some insight into how the ability to have soldiers may have evolved, and revealing that in one species of bee at least, there ARE soldiers!

The PNAS article, from researchers in Sussex (UK) and Sao Paulo (Brazil), studied the neotropical stingless bee Tetragonisca angustula. This species has two kinds of guards that protect the nest: standing bees near the entrance, and ‘hovering bees’ that do pretty much that. Together, only about 1% of T. angustula workers in a nest are guards.

The guards are much bigger (both in size and weight) than their sisters who go out foraging, or those who take the trash out, suggesting that they are actually determined to perform this function (no, bees don’t have free will, either). Furthermore, they are not simply scaled up versions of their smaller sisters, bits of them are proportionately bigger, suggesting that they are specialised in certain functions.

T. angustula forager (left) and guard (right). Grüter et al (2012), PNAS.

Having shown that this species of bee does have ‘soldiers’ – a morphological worker subcaste – the next question was ‘why’? These stingless bees can’t defend themselves by stinging (the clue is in the name), and yet they are regularly attacked by much larger robber bees (the clue’s in the name there, too).

In a final step the scientists carried out a classic experiment: they set up a bug fight. T. angustula guards were paired off with their deadly foes, Lestrimelitta limao robber bees. The larger the guard, the longer the fight lasted.

L. limao bee (right) with the head of a T. angustula guard clamped onto it. The robber bee has decapitated its foe, but has not been able to shake off the head, and cannot fly. Grüter et al (2012), PNAS.

The increased size, and perhaps the altered shape, of the guards, may be involved in their defensive ability. As the authors say in a nice closing sentence: their discovery ‘serves as a reminder that stingless does not mean defenceless’.

In the best scientific tradition, this finding merely raises more questions: if T. angustula has morphological soldiers, why don’t other bee species, or wasp species, for that matter? Is there something different about the ecology of ants and termites (you’ll have noticed they are both terrestrial) compared to that of bees and wasps that means that the evolution of soldiers is more likely?

In some closely related ant species can vary for the presence of soldiers. For example, in the genus Pheidole, although all species have two worker subcastes (minor workers and soldiers) as well as the queen, eight species also have ‘supersoldiers’ – much larger individuals, with significantly wider heads and with tiny vestigial wings. It is thought that juvenile hormone (JH) levels – largely influenced by the amount of food the insect has received as a larva – are responsible for these switches. The presence of apparently useless vestigial wings in supersoldiers is probably due to some developmental link between extra-high JH levels and wing development, which is required for queen development. Don’t forget, all these kinds of ants have the same genes, they are just being activated in different ways at different times in different tissues, due to the amount of JH, which in turn is affected by food.

Three for the price of one: juvenile hormone levels alter morphology in Pheidole ants (adapted from Rajakumar et al (2012) Science

A study in Science by an international group of researchers looked at the evolution of supersoldiers, and how they might develop. The starting point was a chance observation on a field trip, where they found some odd soldiers in a nest of Pheidole morrisi, which does not have supersoldiers.

These odd ants were larger than their sisters (see below), and looked a bit like supersoldiers – right down to the vestigial wings – suggesting that this species might also possess the developmental pathway to produce these extra-large soldiers, even though they are not normally found in nature.

Wild-caught P. morrisi soldier (SD) and anomalous supersoldier (aXSD). Rajakumar et al (2012)Science

By looking at a number of Pheidole species, the authors concluded that naturally-occurring supersoldiers had probably evolved separately, even though similar developmental mechanisms seem to be involved.

To investigate what those mechanisms might be, the researchers put methoprene (an analog of insect juvenile hormone – JH) onto larvae of P. morrisi and produced extra-large ants a bit like the ones they had seen in the wild. They then tried a similar experiment in other species that do not have supersoldiers and found similar results. It seems that all Pheidole ants they studied share the ability to produce supersoldiers, if the right environmental conditions apply.

Evolutionary history of ancestral developmental potential and phenotypic expression of supersoldiers (XSDs). MYA, million years ago. Purple represents the expression pattern of the sal gene in the larval tissues ('wing discs') that will give rise to wings; asterisks indicate the absence of vestigial wing discs and sal expression. Green arrows and boxes represent the induction of XSD potential. From Rajakumar et al (2012) Science

To confirm that JH levels are involved in the production of supersoldiers, they studied a species that has this caste, and put JH on larvae that had yet to ‘decide’ whether they were going to be ordinary soldiers or supersoldiers. The treated larvae produced a higher proportion of supersoldiers, suggesting that JH is indeed involved in determining whether a supersoldier is produced.

This neat study doesn’t resolve the issue of why only some Pheidole species have this extra caste, but it seems very likely to be related to their ecology – some of the species with supersoldiers have to fight off attacks of army ants.

Whatever the case, JH levels, driven by food supply, appear to be responsible for morphological variation in these ants – and, I would bet, in the stingless bee soldiers too. The question of why such variability occurs remains unclear, and probably does not even have a single answer.


References (links to abstracts; pay wall for full articles)

 Rajendhran Rajakumar, Diego San Mauro, Michiel B. Dijkstra, Ming H. Huang, Diana E. Wheeler, Francois Hiou-Tim, Abderrahman Khila, Michael Cournoyea, and Ehab Abouheif (2012) ‘Ancestral developmental potential facilitates parallel evolution in ants’ Science 335: 79-82.

Christoph Grütera, Cristiano Menezesb, Vera L. Imperatriz-Fonsecab and Francis L. W. Ratnieks (2012) ‘A morphologically specialized soldier caste improves colony defense in a neotropical eusocial bee’ PNAS Early Edition.


  1. SmoledMan
    Posted January 10, 2012 at 1:51 pm | Permalink

    You know, just because we can genetically engineer anything doesn’t mean we should. I know these scientists are having an absolute ball playing god and all, but what about ethics?

    • Kevin
      Posted January 10, 2012 at 3:34 pm | Permalink

      The scientists are in no way and in no measure playing god.

      Such a creature does not exist.

      They are, however, playing scientist. And they’re very good at it.

      However, a lot of diabetics, who get genetically engineered insulin, or hemophiliacs, who get genetically engineered clotting factors, would disagree with you.

    • Mary
      Posted January 10, 2012 at 3:41 pm | Permalink

      Any research involving organisms has strict ethical guidelines that must be followed. It’s folly to think that scientists have the freedom to do whatever they want

      • SmoledMan
        Posted January 10, 2012 at 3:43 pm | Permalink

        Why are they creating super-ants if the end goal isn’t to figure out a way of creating supermen? I can see the Eugenics Wars are right around the corner…

        • Ryan
          Posted January 10, 2012 at 3:49 pm | Permalink

          Poe, right?

        • 1000 Needles
          Posted January 10, 2012 at 4:28 pm | Permalink

          The larval development of ants and bees is in no way analogous to human embryonic development. No tinkering around with insect hormones would ever yield a result that is applicable to human body size.

          It sounds like your idea of science was largely formed by the “mad scientist” villains of comic books and movies.

          These scientists are ant and bee specialists trying to learn more about ants and bees. There’s no ulterior motive, except to increase the knowledge in their incredibly niche field.

          • SmoledMan
            Posted January 10, 2012 at 5:32 pm | Permalink

            You say that now, but the truth is their knowledge serves as the base for future research. I can imagine plenty of ruthless governments are not above super-engineering humans.

            • Diane G.
              Posted January 11, 2012 at 2:41 am | Permalink

              See baseball, circa 1990–2005.

            • Notagod
              Posted January 11, 2012 at 8:16 am | Permalink

              So your plan is to stop the open and honest scientists from gaining knowledge and in your mind that will stop the unethical dishonest people from doing research. I think your plan is far more likely to end in disaster because the only ones that will have knowledge are the unethical dishonest people.

        • Posted January 14, 2012 at 12:07 am | Permalink

          Are you retarded? Larger bodies are bigger targets for bullets.

      • Achrachno
        Posted January 10, 2012 at 10:02 pm | Permalink

        Do they need to follow particular ethical guidelines when working with things like ants in an experiment like this? If we allow people to spray and squash ants freely, it seems absurd to make an issue over tinkering with hormone concentrations.

        • Posted January 10, 2012 at 10:13 pm | Permalink

          No,animal care guidelines generally apply for work on vertebrates, not inverts.

          • SmoledMan
            Posted January 10, 2012 at 10:24 pm | Permalink

            Why discriminate against any form of life?

            • Achrachno
              Posted January 10, 2012 at 11:07 pm | Permalink

              I don’t know. Ask your dr. about that the next time you come down with pneumonia or something similar and he wants to give antibiotics.

            • Chris Granger
              Posted January 11, 2012 at 5:10 am | Permalink

              You can’t be serious. Should we start a petition to stop the cruelty to plants when they’re harvested from a farmer’s field? Are you trollin’, bro?

            • Notagod
              Posted January 11, 2012 at 8:23 am | Permalink

              Why don’t you eat yourself instead of other life forms?

    • Ryan
      Posted January 10, 2012 at 3:48 pm | Permalink

      Well good thing neither study genetically engineered anything huh. The species they tested all had the genetic potential downstream to express the soldier caste. They likely differed in their production of JH, what they needed was an external application of that hormonal trigger. Just curious, what on earth could the ethical implications be to this work that you are so concerned about?

      • SmoledMan
        Posted January 10, 2012 at 5:33 pm | Permalink

        Why is man tampering with nature like this? God designed everything perfect already, we have no business questioning HIS design.

        • Persto
          Posted January 10, 2012 at 6:45 pm | Permalink

          Hahaha. You have to be joking?

        • Achrachno
          Posted January 10, 2012 at 10:16 pm | Permalink

          God did a poor job, obviously. No imagination as usual. Improvements are needed.

          It’s gonna be great when we develop 10 ft. long super-soldier ants that will do our bidding!

          But, I really want to see some giant nasutiform termites each with a huge nozzle on the front oozing interesting, customized, bioengineered chemicals. We can train armies of them to go forth and douse anyone who needs it. Heh, heh, heh! Cackle!

          Scared yet, Smoled?

        • Posted January 11, 2012 at 5:15 am | Permalink

          Grow up SmoledMan.

    • Matthew Cobb
      Posted January 10, 2012 at 3:48 pm | Permalink

      No genetic engineering here. Read the post again.

      • SmoledMan
        Posted January 10, 2012 at 3:51 pm | Permalink

        They reactivated a dormant gene using a hormone, you don’t call that genetic engineering?

        • whyevolutionistrue
          Posted January 10, 2012 at 5:35 pm | Permalink

          No, that’s not what genetic engineering is.


        • Posted January 10, 2012 at 5:39 pm | Permalink

          No. That’s not genetic engineering. And you need to read the post again. They aren’t even activating dormant genes.

    • PB
      Posted January 11, 2012 at 12:48 am | Permalink

      i don’t think this article refer to genetic modifications, or creation of super-soldiers to work for super-gaddafies …

      This is a very interesting finding of a specific molecule that create a separate phenotypic differences in larvae of certain insects. This is important, and very interesting scientific fact.

      As about the evil of science and technology – why don’t you ponder the uses of cell-phones as quick remote detonators? … ; )

  2. Kevin
    Posted January 10, 2012 at 3:31 pm | Permalink

    Awesome. Cool science on a crappy Tuesday.

  3. Dan L.
    Posted January 10, 2012 at 4:06 pm | Permalink

    Matthew, wonderful post. Thank you.

  4. Posted January 10, 2012 at 4:45 pm | Permalink

    “Things get even odder when you realise that although many ant or termite species have ‘soldiers’, there are no known cases of bees or wasps having morphological worker casts. There are no ‘soldier bees’ or ‘soldier wasps’. Why not?”

    Perhaps because there is more occasion for warfare on the ground than in the air?

    A quibble: bees have castes, worms leave casts.

    • Posted January 10, 2012 at 5:47 pm | Permalink

      If stingless ants and stingless bees have soldiers maybe it’s the stinging. Do stinging ants, like bull ants, have soldiers?

    • PB
      Posted January 11, 2012 at 12:49 am | Permalink

      .. because dictators and super-gaddafies are terrestrial ? ..

  5. Posted January 11, 2012 at 5:12 am | Permalink

    Ants make blankets!

  6. Blondin
    Posted January 11, 2012 at 6:38 am | Permalink

    Aaiiiiii! It’s the Zanti Misfits!

    • horrabin
      Posted January 11, 2012 at 10:59 am | Permalink

      “Total destruction to those who would invade our privacy!”

  7. Achrachno
    Posted January 11, 2012 at 7:50 pm | Permalink

    “there are no known cases of bees or wasps having morphological worker casts.”

    Colonial bees (honey bees, and some others) have workers, drones and queens, so that’s not quite right. The workers do not reproduce and are morphologically different from the reproductives.

    I believe all wasps form colonies of equals, and all individuals are reproductive. That would tend to prevent the formation of specialized soldier castes, I would think.

    Likewise many “solitary” bee species consist only of normal reproductive individuals and there are no specialized worker castes.

    Someone can correct me if I’m wrong. There may be a wasp species somewhere that makes sterile workers, but I’ve not heard about it.

4 Trackbacks/Pingbacks

  1. […] Soldier bees, super-soldier ants | Why Evolution Is True. Like this:LikeBe the first to like this post. […]

  2. […] Soldier bees, super-soldier ants ( […]

  3. […] Soldier bees, super-soldier ants « Why Evolution Is True. Share this:TwitterFacebookLike this:LikeBe the first to like this post. By Colin Mackay 0 […]

  4. […] Soldier bees, super-soldier ants ( […]

%d bloggers like this: