You lazy, good-for-nothing thrips

I thought I’d share some recent and preliminary data from my thrips, which should give a flavour of what I’m doing in the outback.  These are quite tentative, and I’m hoping to increase samples next time I go out.

Dunatothrips build silk nests (see our recent paper), and there are usually between 1 and 4 adult females inside a nest (although this can get up to 20).  If I experimentally damage the nest, the females repair it.  You can see this in this video I took down the microscope:

Notice that one female remains in the background, doing nothing, while the others work hard to add silk to the nest.  I saw quite a few females refusing to help in this way. I recorded which of the females chose to engage in what I assume is risky and costly repair behaviour. I removed the females in the order they began adding silk to the nest, and classified them into “repaired” versus “did not repair”.

Then, I dissected them to see whether they were reproductive (i.e. whether they had developing eggs in their ovaries). I found that females that chose to repair the nest were usually reproductive, while those that did not repair were non-reproductive (see below – I’ve applied some scatter to the points to make the figure clearer).


(The bars show the probability of me finding the female has developing ovaries when I dissect her, given what I have observed of her behaviour).

Thus, some females are neither reproducing nor helping to maintain the nest.  Sounds a bit like human teenagers to me.  Except these are not offspring (the nest hasn’t matured yet, and nests only support one generation of offspring).  They are most probably sisters or near relatives of the reproductives (see this paper here for evidence).

Can we say anything else about these do-nothing thrips at this stage?  I have also found (very tentatively) that you tend to get nonreproductive females in small nests, while in larger nests, everybody reproduces:

prop repro vs volume

If this indeed turns out to be true, maybe the size of small nests stops some females reproducing.  This would make sense, because there is only a limited surface area of leaf to feed on inside a nest.  If we look at the wider sample of nests I’ve collected on all my field trips combined, we can see the pattern of offspring production also supports this idea.  In large nests, more females produce more total offspring.  In small nests, more females do not produce more offspring.

total offspring vs nest size

This suggests that in small nests, some females are neither having their own offspring nor helping their nestmates produce more offspring (in both cases, we would have seen increases in total offspring with numbers of females).

Of course, “lazy” nonreproductives may be helping in ways I can’t detect – perhaps by helping the offspring to feed, by defending against an enemy I haven’t encountered yet, or by maintaining nest hygiene.  In those small nests, the offspring may still SURVIVE better if there are more adult females in the nest.  That’s something I’m going to have to test, maybe by measuring offspring size, or development rate in the field. Alternatively the reproductives might survive better if they don’t have to work as hard when helpers are present, something that would certainly be difficult to test in the field.

The big question is, what determines who gets to reproduce in a small nest?  Intuitively, there should be competition over breeding status, as happens in (for example) paper wasps and joint-nesting ants.  As we described in our last paper, these thrips appear to be pacifists, with no apparent aggression at all.  Also, there’s no evidence that nonreproductive females are smaller than reproductive females from the same nest.  Unless they are conducting some kind of chemical warfare, there is no indication of conflict at all.

In other species, lazy workers are sometimes waiting for a chance to reproduce themselves – the more they work now, the less energy they have for later reproduction.  In Dunatothrips, this is probably not the case because nests probably only have one generation. However, that is not set in stone, and it may be that females in a small nest end up taking turns to reproduce in what we would call a social queue. That would be really exciting because it would make this system quite close to Polistes paper wasps.  These are questions I’m hoping to solve with experiments next time I hit the field.  I am currently thinking about how to design these, and constructive ideas are most welcome!

Associating with ants: a dangerous game

Spiders take a leaf out of Spielberg’s book in dealing with potentially deadly adversaries, according to recent research.  “Keep absolutely still.. his vision’s based on movement.

What if you had an unassailable personal army of bodyguards, but who would kill you the instant they knew they could?  What if your only available prey were potentially vicious killers?  You have to be able to catch one, and at the same time avoid becoming an hors d’oeuvre on *their* menu.

Such is the problem for any arthropod that decides to rely on ants for its survival.  To put it mildly, they’re not the world’s easiest, plumpest or least-resisting meal.  They can get a bit, shall we say, argumentative.

But if you can crack it, boy, you’ve got it made.

Plenty of insects have given this a shot – so many that the habit is given its own name, “myrmecophily”.

For example, the butterfly family Lycaenidae have this down to a fine art (explored in this review by Pierce et al 2002).

Liphyra larva. Photo: © Dani Jump
Liphyra larva. Photo: © Dani Jump

Take Liphyra, which I remember most clearly because it was featured on one of David Attenborough’s marvellous programs. The caterpillars are shaped like impregnable tanks and slowly gorge themselves on ant larvae while the ants are helpless to stop them.  Once emerged, the adults are covered in waxy scales that the ants cannot grip, and waltz out of the nest to fly away unharmed.

Some Lycaenids have evolved glands (called “cupola glands”) that secrete perfume that mimics the ants’ own secretions.  Consider for example the caterpillars of Niphanda, which cloak themselves in the perfume of ant larvae, and then sit back and relax as they are enthusiastically fed to adulthood by the ant workers.  Hojo et al (2009) found that the ants would even feed glass beads that had been experimentally rolled in the correct perfume.

Niphanda larva being fed by ants as one of their own. Photo:
Niphanda larva being fed by ants as one of their own. Photo:

Beetles of the tribe Paussini have arrived at a similar set of solutions, reviewed here: some are shaped like little tanks while others cloak themselves in protective chemicals.  In all cases the beetles munch happily away on the ant larvae while the ants can do nothing to stop them.

Spiders can also use chemical trickery: Cosmophasis spiders live within ant nests, again safely cloaked in “Eau de Fourmi” acquired (in a macabre twist shown by this study) directly from the ant larvae on which they feast.

Aphantochilus spiders, although they are not chemically or physically protected, have evolved a very particular set of skills to allow them to pick off individual ants. They are still vulnerable to attack, though, especially when small. Accordingly the mother defends small offspring vigorously against attacks by their future prey.

Protection rackets

Many arthropods use ants as protection.  No-one’s likely to attack you if you are surrounded by thousands of angry, biting micro-warriors.  But the same principle applies; nothing’s stopping those warriors from turning on you.

Aphids, treehoppers and caterpillars provide “protection money”.  They are well-known to be farmed by ants, a great example of a protection mutualism where the insect produces a sticky, sugary “gift” in return for the ants’ protection.  Treehoppers that normally provide parental care known to turn over responsibility for their offspring entirely to ants.  This recent study found that treehoppers can even summon ants using signals made by vibrating their wings when they are attacked by predators.  These associations run so deep that ants will even protect aphid eggs, which do not produce honeydew – in effect the ants are investing effort for a future reward.

Sweet rewards are cheap for aphids and treehoppers (since honeydew is waste material to them) but caterpillars must produce this substance de novo, making it rather expensive.  Lycaenids do this by means of “tentacular glands” (see Pierce et al‘s review).

Extra reading about mutualisms with ants can be found in this chapter of the wonderful book Ant Ecology (OUP, 2010).

The old Jurassic Park trick

Eustala spiders are not so generous.  A new study in the journal Ecological Entomology by researchers Loriann Garcia and John Styrsky shows that Eustala like to keep it simple; no complex chemicals, super-thick armour, or twinkle-toes defence techniques – and definitely no sugary gifts for the ants.

Instead, like that nerve-jangling scene from Jurassic Park, what Eustala does is to stay really, really, really, really still.  Even with ants crawling all over it, feeling and probing it with their antennae.  And the ants don’t notice it.

Keep absolutely still, his vision's based on movement.. Photo:
Keep absolutely still, his vision’s based on movement.. Photo:

By day, Eustala spiders rest on Acacia trees in Panama.  They sit on the thorns, and oddly are not attacked by the ants that swarm over the plant and over the body of the spider.

Why are they not attacked?  The authors did an experiment using both Eustala and Argiope, a spider that lives on ant-free trees.  They placed recently-killed (frozen and just-thawed) spiders of both species onto branches containing ants, and watched what happened.  For comparison, they released moving spiders (still agitated after being handled) of both species onto nearby branches.

Playing sleeping lions… with actual lions

Regardless of species, the immobilised spiders were not attacked, but the moving spiders were.  Moving

Eustala hiding in plain sight. Photo © Loriann Garcia, John Styrsky
Eustala hiding in plain sight. Photo © Loriann Garcia, John Styrsky

Eustala had no particular advantage over moving Argiope.  That shows that Eustala is not primarily relying on any kind of chemical camouflage to defend against the ants.

Furthermore, Eustala had one more trick up its sleeve.  If discovered, Eustala jumped to safety on a line of silk, allowing them to cautiously rappel back up the line once the ant attack subsided and resume their position.  Once in this position, they did not react at all to patrolling ants, even when tickled by ants’ antennae and lunged at by ants’ jaws.

Argiope, on the other hand, tended to try and run away, which further enraged the ants and ended up with them being killed or driven off the plant entirely.

One huge Tyrannosaur or millions of tiny ones: clearly the advice is the same.

Termites measure wood from inside using echolocation

C. secundus worker chewing a piece of wood. Image: © Simon Tragust and Tobias Weil/DDP
C. secundus worker chewing a piece of wood. Image: © Simon Tragust and Tobias Weil/DDP

You’re sitting in a cave.  How big is the mountain above you?

A termite would be able to tell you, simply by chewing the wall.

Drywood termites live quite bizarre lives when you think about it.  Next time you hear the loud noises of termites chewing inside a wood block, take a moment to think about how odd their existence must be.

Imagine being completely entombed in a cave made of delicious ham. You live in it, tunnel through it, raise your kids in it, excrete in it.

Wings and eyes are both useless, so you have dispensed with them.

At some point down the line, one generation of your unborn great-grandkids is going to find the ham runs out.  And that’s a problem. They are going to have to fly off and find another mountain of ham that can sustain them and their offspring.

But – and here’s the spicy port-and-molasses rub – you have to know how much ham is left  in order to know when to develop wings and fly away rather than stay and have babies.  And when you find a new mountain of ham, in order to know whether it is a suitable resource for you and your future kids, you have to be able to assess how big it is.

From the inside.

How do termites know how big the wood block is without measuring it?

Some insects walk up and down a resource to measure it, like burying beetles, which walk round a dead mouse several times before deciding whether to breed on it.  That doesn’t work for termites, because predators would quickly snap them up.

An experiment a few years back discovered that Cryptotermes termites actually use a form of echolocation to size up the piece of wood they’re in – and to choose between alternative blocks of wood.

Staggeringly, the termites can sense the vibrations that are caused when they make their loud chewing sounds, and the resonant frequency tells them how big the wood is.

Faced with a choice of 20mm and 160mm blocks of pine, Cryptotermes invariably chose the smaller block (probably because they are small-bodied termites and don’t want to bump into any larger ones).

But when chewing vibrations from the 160mm block were recorded and then played back through the 20mm block, the preference was destroyed.  When control vibrations consisting of random noise were played through the 20mm block, there was no effect on preference.

Other insects communicate using vibrations through wood or other material – for example, treehopper babies “buzz” against their branch in structured collective patterns, using a “cooler!  warmer!” method to direct their mother to the source of a predator attack. Passalid beetle larvae communicate with parents via cicada-style stridulations, while some male dung beetles produce a courtship song that is transmitted by vibrating the poo in which the female is buried.  They say romance is dead…

On cannibalism and spider bondage

Tips for males to avoid being eaten during sex

Redback male cautiously approaching a female. Photo:Daily Telegraph

 “Love hurts”, according to Roy Orbison, Aerosmith, and too many others to list.

A clichéd metaphor for us – but in some species, it is a bit too true for comfort.

One of the weirdest observations in nature, for example, has to be the sexual behaviour of the Australian redback spider.

In the midst of mating (which, by the way, happens between organs on the male’s mouth and the female’s belly, as in all spiders) the male redback spider does a spectacular reverse somersault, using his mouth-based genitals as a pivot, and impales himself on the female’s jaws (see this video from about the 2:00 mark).  While he finishes what a man has to do, the female sucks his body dry.  Up to seven out of ten males die this way, during sex.

Being cannibalized isn’t so bad

Praying mantis female chews off the head of a male. Photo:Biozine

I would hope by now your scientist brain is screaming “WHY?!” Surely Darwin is turning in his grave?

Darwin would probably approve, as it happens.  Male redbacks that commit suicide in this fashion manage to mate for longer and fertilize more eggs than those that do not. And it’s not just redback males that are perfectly happy to lie back and think of Australia while being horribly eaten.

Cannibalized male Hierodula mantises can take comfort from the fact that their flesh-fed mates will have more babies than those of uneaten males.

A male Nephila spider can feel assured that his shriveled, torn-off genitals will stay attached to the female for hours afterwards, repelling other males.

As if this were nothing, Argiope male spiders spontaneously die in the act of sex, leaving their dead body as an obstacle to future amorous males. Face it, removing the half-eaten corpse of a previous lover is not really a turn-on.

50 reasons to eat your lover

What does the cannibal female get out of it?  “Man-eater” females are not rare among arthropods – especially among praying mantises and spiders, whose females are often observed chowing down on hapless would-be hubbies.

In many cases, the hungrier a female is, the more she is likely to eat the male, even though males often make poor-quality meals.

By attacking everybody, female orb spiders can weed out the weaklings and only mate with the strongest males.

Female wolf spiders cannibalize no-hoper males whose attention just attracts predators.

But why are mantises and spiders special? Shawn Wilder and colleagues have found that size difference is one key – the smaller males are relative to females, the more likely they are to be gobbled up.  In mantises and spiders, females commonly dwarf males.

The self-preservation society

But not all males go quite so gently into that good night as do male redbacks.  Understandably, males of most species would rather not be cannibalized. And faint heart does not win fair lady. According to this recent study, male mantises are more willing to risk being killed and eaten if females are scarce in the environment.

But a male on the pull can employ a number of tricks to avoid ending up on the menu.

Tenodera mantises have an obvious tip: only chat up a cannibal when she has just eaten.

Male Pisaura spiders go one further and offer her dinner first before trying it on (although occasionally the female still thirsts for blood, whereupon they feign death).

Pisaurina males tie females up before mating. Source:Bruce & Carico (1988)

For some male spiders the solution is to get kinky. The male Pisaurina spider ties his mate up in silk before sex, to make absolutely sure she can’t eat him.

Male Agelenopsis are into asphyxiation games – the male gasses the female with a narcotic pheromone, ensuring she is too groggy to attack.

Some males have an insurance policy. In the 1930s it was discovered that male praying mantises have a second, extremely horny brain in their genitals that takes over if the main brain is eaten. As Greg Holwell explained at a recent conference, this brain is capable of steering the genitals to the appropriate bits of the female even when the male is nothing but a disembodied abdomen. As soon as the female eats the male’s brain, the sex-crazed backup takes over and immediately lunges for her privates.

The courses of true love

Finally, some males have turned the tables and indulge in a little role reversal.  According to this study, if a female wolf spider’s looks and sexual history are not up to scratch, it is she that is apt to become lunch for the hungry male.

The take-home message is this: when a pop star warbles your way about the pain of love, at least be glad you’re not an spider or a mantis.

I’m running the Sydney Marathon

An admittedly un-entomological side note, but worth plugging nonetheless.  In 6 weeks I will be running the Sydney Marathon in aid of one of the most honest and worthy causes possible – my Mum’s amazing charity supporting Bedouin in the South Sinai, Egypt.  If you’re interested to know more please check out our Aussie blog for details, or go straight to my JustGiving page. If you’re in the UK, you can donate with just a text message.

Monster-eyed bugs

Silkworm moth showing eye spots

Eyespots probably are eyespots after all.  Moths really do try to fool birds into thinking they are staring at an angry cat or owl using fake “eyes” on their wings, a new paper shows.

The researchers cleverly manipulated the position of the white “sparkle” in the moths’ fake eyes.  If the “sparkle” appeared in an unrealistic position, the moth was eaten.

Moths and caterpillars often display pairs of large dark circles.  Tradition has it that these represent eyes – they are even called “eye spots”.  But are these really meant to be eyes?  Research shows that the patches cause predators to “startle”.  However, there has been a recent blazing row minor academic debate over whether these really are images of eyes.

On the one hand, many other moths achieve a “startle” response by flashing bright patches of colour, and the spots may be nothing more than brightly coloured patches.  For example, one recent paper showed that birds avoid moth-like dummies with square or triangular patches just as much as they avoid dummies with circular patches, and that stripes are just as good protection as spots.  Thus, it may simply be their conspicuousness that protects the moth, not their resemblance to eyes.

Classic eye from anime, showing highlight.

But an ingenious new study out last week has demonstrated that most moths with “eyespots” in fact use a clever illusion that makes it very hard to believe these are not meant to be images of eyes.  Most “eyespots”, the authors found, have a “sparkle” – an off-centre white patch that looks like the highlight reflected by the surface of an eye.  All cartoon eyes have a “sparkle”, for example – a simple way to enhance their realism.

Selection of eyespots from moths

Graph showing number of moths eaten with sparkles in different places.

Not only did the authors discover that almost all eyespots carried by moths have this sparkle, but they demonstrated that the position of the sparkle has a clear effect upon the moths’ survival.

What makes this experiment doubly clever is that, by manipulating only the position of the “sparkle”, the researchers cleverly kept the spots at exactly the same brightness, but changed how realistically they resembled a real vertebrate eye.  The fact that this still affected how quickly the moth was eaten shows that the eyespots really are meant to resemble eyes.

All of us at some point have jumped sky-high at what we think is a tarantula crawling up our leg, which in fact turned out to be a piece of fluff.  What we experience here is a startle response, triggered by something resembling a large, hairy spider.  This kind of reaction to potentially dangerous animals makes a lot of evolutionary sense.  Until we realise it’s a piece of fluff, the only thing we want to do is get away from it.  If this were a piece of food – however delicious it might actually be – the last thing on our minds would be pecking at it.

This is what some moths achieve by carrying around the image of two large eyes on their wings.  For a moment, the bird believes it is staring an angry cat in the face, or an owl.  Superb!