This spider pretends to be an ant, but not well enough to avoid being eaten

If Spiderman and Antman took their DNA and mixed it together in a petri dish, the result might be something like the spider species Siler collingwoodi (S. collingwoodi). This tiny, colorful, jumping spider found in China and Japan uses a combination of camouflage and some award-worthy mimicry to deter some hungry predators. In a stressful scenario, these spiders will imitate the way an ant walks to avoid being eaten.

[Related: Black widows battle their even deadlier cousins in a brutal spider war.]

A study published May 17 in the journal iScience found that the combo of camouflage and ant mimicry works to evade spiders that eat other spiders, but not hungry praying mantises. It’s advantageous to mimic an ant because they are typically not very tasty, and can have spiny defenses, chemical repellents, or venom. Not to mention, species of “exploding” ants like Colobopsis saundersi that are not afraid to fight and bite back. While scientists already knew that S. collingwoodi walked like an ant, the team on this study were curious how accurate the mimicry is, whether it imitates multiple species of ants, and how effective it is at discouraging predators. 

“Unlike typical ant-mimicking spiders that mimic the brown or black body color of ants, S. collingwoodi has brilliant body coloration,” co-author and Peking University in China ecologist Hua Zeng said in a statement. “From a human’s perspective, it seems to blend well with plants in its environment, but we wanted to test whether their body coloration served as camouflage to protect against predators.”

To better understand how these ant-inspired theatrics help the spiders avoid becoming dinner, the team collected wild ant-mimicking spiders from four spots in southern Hainan, China, and brought them back to the lab. They also collected another type of jumping spider that does not mimic ants as a comparison and five co-occurring ant species as potential models.  

The team then compared and characterized how the insects and arachnids moved in terms of how they used their individual limbs, their speed, acceleration, and whether they followed a straight path or took a more roundabout way. 

Inside of jumping like most jumping spiders, S. collingwoodi scuttle around like ants. They raise their front legs to mimic an ant’s antennae, bob their abdomens, and lift their legs to walk more ant-like. Out of the five ant species studied, the spider’s style of walking more closely resembled three of the smaller ant species that are closer in size.

“S. collingwoodi is not necessarily a perfect mimic, because its gait and trajectory showed high similarity with multiple ant species,” said Zeng. “Being a general mimic rather than perfectly mimicking one ant species could benefit the spiders by allowing them to expand their range if the ant models occupy different habitats.”

Then it was time to test these defenses against two likely predators. Portia labiata and the praying mantis. Portia labiata is a similarly sized jumping spider with color vision who specializes in preying upon other spiders. The praying mantis is a more generalist predator that has a monochromatic visual system–meaning it has trouble telling multiple colors apart. 

[Related: Jumping spiders might be able to sleep—perchance to dream.]

To see how the color camouflaging was working, they modeled how the two predators would perceive S. collingwoodi relative to the other prey species. They used a background of two plants that the spiders live on—the red-flowering West Indian jasmine and the Fukien tea tree The ant-mimicking spiders were better camouflaged from both predators on the jasmine plant than on the tea tree plant.

The predators were more likely to attack the non-mimicking spider than the ones that imitate ants. Out of 17 trials, the spider launched five attacks—all of them were launched towards a non-mimicking spider. However, praying mantises attacked both prey species with equal readiness.

“We initially thought that both predators would behave similarly in the anti predation experiments, but in fact the simulated ant locomotion of Siler collingwoodi only worked for the jumping spider predator, while the praying mantis showed indiscriminate attacks on both ants and mimics,” co-author and Peking University evolutionary ecologist Wei Zhang said in a statement. 

It is possible that this difference might be driven by each predator’s likelihood of being injured if they eat an ant. The praying mantises are much larger than their prey, and they have a better chance of eating spiny ants without risking catastrophic injury. Predatory spiders do not have this margin for error. 

“For the spider predator, a random attack on an ant could result in injury,” says Zhang, “so they are very careful predators and will only attack if they can distinguish S. collingwoodi from ants with a high degree of certainty.”