Ant- and wasp-mimicking jumping spiders

July 20, 2021 • 12:30 pm

For today’s biology lesson as I get my teeth cleaned, here’s a 20-minute video lesson about some salticids—the family of jumping spiders—that mimic ants. As you’ll see, this resemblance appears to be a form of Batesian mimicry, in which the spiders mimic toxic, unpalatable, or dangerous ants. The remarkable near-perfection of this mimicry, in which many features of the spiders have been modified to look like ant features, shows how closely natural selection can take an organism to its “optimum” phenotype. (Mimicry is one of the feat aspects of organisms in which you can judge how close they come to the “optimum”: in this case the organism or aspect of the environment they’re imitating.) It also shows the ubiquity of genetic variation, which must, during selection, have been present for every one of the modified features.

And it’s not just morphology that gets imitated, either. There’s chemical mimicry in this case, and behavioral mimicry (e.g., how the spider holds its legs in an antlike position).

The YouTube notes say this:

An exploration of jumping spiders that mimic ants (aka Ant-Mimicking Jumping Spiders) framed around a discussion with spider-scientist Alexis Dodson of the University of Cincinnati’s Morehouse Lab. 

h/t: Jacques

Readers’ wildlife photos (and video)

July 23, 2018 • 7:45 am

First a video from reader Rick Longworth, who added these notes: “This spider had just finished molting and was resting under its discarded exoskeleton. The whole process took about 30 minutes of drying as the sun was setting. The tight new outfit she’s wearing gets a bit of scratching at the base of the legs. The light color, I suspect will gradually darken.”

I’m not sure what species this is, but I’m sure readers can help:

And some golden ants and other insects (as well as ant-mimicking jumping spiders) from reader Tony Eales in Oz. His notes are indented:

I recently came across a nest of Polyrhachis rufifemur, which is a spectacular golden species of Tropical Spiny Ant. That lead me to this paper on the golden ant mimicry complex. Over the years I have managed to photograph several instances of this mimicry in ants, wasps, bugs and spiders. Here are a few examples.

JAC: I suspect this is an example of a Müllerian mimicry ring, in which a number of dangerous or toxic species mimic each other, although some of the mimics could be Batesian, meaning that they are harmless and enjoy protection by mimicking a dangerous model. The gold color is, to my knowledge, almost unique as a form of warning coloration, and some of these insects (particularly in the very last photo) are spectacular.)

A velvet ant (Bothriomutilla sp.), which is a wingless female wasp in the family Mutillidae:

A couple of Lygaeid Seed Bugs Daerlac sp.:

Some ant-mimicking jumping spiders Myrmarachne sp.

And ants from diverse families in the genera CamponotusDolichoderusMyrmecia and Polyrhachis (in order):

Looking at the paper, I’ve still got a long way to go to get all the species in this mimicry ring.

A chemically-camouflaged frog

December 1, 2015 • 1:15 pm

By Matthew Cobb

Social insect colonies rely heavily on chemical signalling to identify members of the colony, and conversely to detect intruders. These communication systems are generally very effective, but as that great scientist Professor Ian Malcolm put it, ‘Life will find a way’. If there’s a locked system, somewhere a pesky but perspicacious parasite will find a way to crack it.

Caterpillars of the Maculinea genus – also known as Alcon butterflies – hatch on the ground near Myrmica ant nests, and are picked up by the workers. The ants take home what tastes/smells like one of their babies, except this is a carnivorous cuckoo that will munch its way through their larvae…

An unusual example of such chemical camouflage was discovered in 2013 by a group of German and Swiss researchers, led by Mark-Oliver Rödel of the Museum für Naturkunde Berlin. It’s unusual because, as the title of this post indicates, it involves a frog.

The West African Rubber Frog (Phrynomantis microps) is found throughout west Africa, and can often be found living in the underground nests of the ponerine ant Paltothyreus tarsatus (aka the African stink ant).

These are pretty aggressive ants about 2 cm long, which pack a nasty bite and an even more powerful sting. They can be found pretty much throughout sub-Saharan Africa, where they play a very important ecological role. Most ponerine ants have quite small nests of a few hundred individuals, but in a paper published in 2013 my pal Christian Peeters found that some P. tarsatus nests can be as large as 5,000 individuals.

Here’s a rather small picture of Christian with a box of these alarmingly large ants:

Although the ants are notoriously aggressive, they don’t seem to bother about the Rubber Frog, as shown by this photo:

You can see how the ants seem rather bemused by the frogs in this video stitched together from the Rödel paper by a YouTube user – the first part shows ants with an adult frog, the final section with a froglet:

Other frogs, and other arthropods, are immediately attacked by the ants when they encounter them. However, when dead mealworms or live termites were covered in extracts from the frog’s skin, they were generally ignored by the ants, or at least it took much longer for the first bite to be administered:

Figure 1. Time from first ant, Paltothyreus tarsatus, contact with termites (left; inlet A) or mealworms (right), coated with the skin secretion of Phrynomantis microps, until stinging (inlet B).
Figure 1. Time from first ant, Paltothyreus tarsatus, contact with termites (left; inlet A) or mealworms (right), coated with the skin secretion of Phrynomantis microps, until stinging (inlet B). Control groups are termites or mealworms coated with water. Boxplots show the median and the interquartiles of time from first ant contact with a termite or mealworm until stinging. Coated insects were stung significantly later than control insects. Taken from here.

When Rödel’s group examined the chemical composition of the frog’s skin, they found it contained two novel peptides – short proteins, each 9 or 11 amino acids long – with a proline-phenylalanine pair at the end. When termites were covered with either or both of these peptides, the ants took significantly longer to attack them, suggesting these are indeed the active ingredients on the frog’s skin:

Figure 3. Effect of the two peptides from the skin secretion of Phrynomantis microps applied to termite, Macrotermes bellicosus, soldiers and delaying the aggressive behaviour and stinging of Paltothyreus tarsatus ants.
Figure 2. Effect of the two peptides from the skin secretion of Phrynomantis microps applied to termite, Macrotermes bellicosus, soldiers and delaying the aggressive behaviour and stinging of Paltothyreus tarsatus ants. Maximum observation time was 20. Taken from here.

This finding is doubly surprising – most instances of chemical camouflage involve cuticular hydrocarbons, which many arthropods use for communicating (for example, these are involved in the case of the Alcon Blue caterpillars described above). In the case of Phrynomantis microps, not only were novel peptides involved, no hydrocarbons could be detected on the frog’s skin, even though the animals were living in a hydrocarbon-rich environment in the ants’ nest.

What’s in it for the frog? Protection from predators (you’d have to be very foolhardy to take on the ants) and possibly protection from dessication during the dry season. They may also eat some of the ant larvae, although that is speculation on my part.

What’s in it for the ants? Probably nothing. If the frogs found a way to hack their chemical communication system, but at low or zero cost to the ants, then it won’t matter. If there’s a substantial cost to the ants, then you would expect a chemical arms race to begin – any ant nest that used a slightly different system of communication would not sustain the cost of the frog in the room.

The final point about this rather neat piece of biology, which flowed from a field observation, is that it’s opened up a new area of study in chemical communication in ants, and potentially a way of placating aggressive insects.

You see, Professor Malcolm was right:


Rödel M-O, Brede C, Hirschfeld M, Schmitt T, Favreau P, Stöcklin R, et al. (2013) Chemical Camouflage– A Frog’s Strategy to Co-Exist with Aggressive Ants. PLoS ONE 8(12): e81950

Peeters C, U. Braun U & Hölldobler B (2013) Large Colonies and Striking Sexual investment in the African Stink Ant, Paltothyreus tarsatus (Subfamily Ponerinae) African Entomology, 21(1):9-14. (Abstract)

Spot the Nilgiri Marten!

September 21, 2015 • 1:57 pm

by Matthew Cobb

Aditya Gangadharan (aka @AdityaGangadh) posted this pic on Twitter. To give you a hint, a Marten is a mustelid (methinks it is like a weasel), about 50 cm long, with a tail that is slightly shorter. So we ain’t talking nightjars! Click twice to embiggen sans book adverts, comme d’hab’.


Wikipedia claims:

Very little is known about the Nilgiri marten. It is diurnal, and though arboreal, descends to the ground occasionally. It is reported to prey on birds, small mammals and insects such as cicadas.

I was partly attracted to this photo because the Nilgiri Marten is found in the Nilgiri hills in south-western India. For several years, together with ant-expert Christian Peeters, I worked on a population of ant from this region, called Diacamma. Diacamma ants are queenless – they have lost the queen caste, and one worker dominates the others. She is called a gamergate (pronounced gammergate – the word means, or perhaps meant, given that there’s now another meaning, ‘married worker’) and lacks the large ovaries and wings typical of queens.

This situation is typical of many Ponerine ants, but what is weird about Diacamma is that they possess external glands called gemmae – apparently based on wings – that enable them to produce pheromones and to mate. The dominant female hangs around the cocoons waiting for the new ants to hatch out, and she then bites off their gemmae, effectively sterilising her sisters (and later, her daughters). (We studied how this takes place – contact me if you want a copy of the article, as the full article is behind a paywall.)

When the gamergate dies, or the colony splits and half the colony no longer has a gamergate, the first female to hatch out is not mutilated (there’s no one there to do it), so she becomes the new dominant. For pictures of the gemmae go here.

This situation exists in all Diacamma ants, except the Nilgiri population, where no mutilation takes place and the gamergate imposes her dominance physically, as in other Ponerines. This group is not a separate species from the local D. ceylonense, as the two taxa will interbreed (we succeeded in doing this in the lab – a big deal if you work on ants). You can read an abstract from a recent paper by Christian Peeters here.

The scrappy ground seen in the photo is the kind of place you get Diacamma ants – there might even be some mooching around…