If it looks like an ant and walks like an ant, it must be an ant, right? Thanks to evolution, this isn’t always the case. Plants and animals can evolve to mimic other species in appearance, behavior, sound, or smell. By doing so, mimics can reap benefits such as increased access to food, enhanced reproduction, or a smaller chance of being eaten. For example, tiger moths emit ultrasonic sounds as warnings of their toxicity to bat predators. Non-toxic pyralid moths produce similar sounds to evade bat predation without having to produce the toxins themselves (1). These mimics are nature’s “sheep in wolves’ clothing” because they falsely appear to be dangerous. And because the most effective mimics are less likely to be eaten, they are able to pass their traits to the next generation, meaning that natural selection favors the evolution of more and more convincing mimicry.
For jumping spiders looking to avoid becoming dinner for birds, lizards, and bigger spiders, mimicking an ant might be an effective strategy. Ants have stingers, taste bad due to a range of chemical defenses, and often hang out in groups, meaning that if a predator messes with one ant, it must reckon with a whole colony.
So how exactly does a spider mimic an ant? Ant-mimicking spiders usually have a constriction on their first body segment, which makes it look like two, giving them the overall appearance of three body segments like an ant. Then, they have to make their eight legs appear as only six. To complete their ant appearance, spiders need to gain some antennae too. Some researchers hypothesized that ant-mimicking spiders walk on six legs and hold the front pair of legs ahead of them to appear like antennae. In other words, it was thought that the mimic spiders don’t just look like ants, but also act like them. But this idea had not been tested scientifically until recently. In fact, mimicry of dynamic traits like locomotion have been less well-studied than other characteristics like color or shape (but see (2-4)) . Historically, studies may have been hampered by lack of access to high-tech equipment such as video cameras and computers which allow researchers to quantify dynamic characteristics that are unobservable to the human eye.
A team of researchers at Cornell University set out to investigate the mimicry of ant locomotion in the jumping spider Mymarachne formicaria. Led by Paul Shamble, they were interested in understanding what it means to walk like an ant, what features define how an ant moves, and to what degree spiders duplicate these features. To confirm that the spiders’ motions are truly ant mimicry, the best test is to measure how a predator responds to mimic spiders, non-mimic spiders, and ants. If predators attack the mimics less often than the non-mimics, this means that the spiders’ deception is effective, and is strong evidence ant-like movement evolved to avoid predation.
Shamble and his colleagues collected specimens of two ant species to identify the features of ant locomotion. They also obtained individuals of the ant-mimicking species, and of a different, non-mimicking jumping spider species to see if the behaviours they saw in the ant mimic species were related to mimicry or were just jumping spider characteristics.
High-speed video was used to analyze the animals’ gait as they walked across a large, featureless arena constructed in the lab. By carefully tracking the movement of each animal’s legs, the researchers found that both the mimic and non-mimic spiders walked on all eight legs, contrary to what had been previously suggested. However, when the mimicking spider paused, it raised its first pair of legs upwards to create the illusion of antennae, unlike the non-mimicking spider. Although the mimic did not perfectly replicate an ant’s gait, its behaviour is different (and more ant-like) than the non-mimic spider. More studies on the gait of different ant-mimicking jumping spiders are needed to see whether they all walk on eight legs, or if some spiders have evolved to walk on just six like an ant.
The team also recorded the overall movement patterns of the animals as they walked across the arena. When ants follow chemical (pheromone) trails produced by other individuals of their species, they walk in a characteristic wave-like path; if no pheromone cue is present, ants walk in a random winding pattern. The non-mimicking spider walked in a relatively straight line, while the mimicking spider’s trajectory was wave-like, similar to that of ants following a pheromone trail. The mimicking spider’s movement was indistinguishable from that of ants, providing further evidence that these mimics really do behave like ants.
Once the researchers had demonstrated that the movement of the mimic spider was similar to the ants’, the next step was to test whether the mimicry was effective enough to fool a large predatory spider. Shamble and his colleagues used simplified video animations of the animals, manipulating movement and gait to study how varying these traits altered attack rates. The predatory spider readily attacked video animations of the non-mimicking spider. However, it was much less likely to attack the animations of ants or ant mimics, confirming that the mimicking spider’s ant-impersonation was effective in reducing number of attacks.
The existence of locomotive mimicry has often been suggested, but research that verifies it is rare. This study is important because it provides quantitative evidence that the predator spider was duped by the mimic’s ant imitation. In the wild, reduced predation of good mimics means that they are more likely to survive and reproduce, passing on the genes that underlie ant-like movement to their offspring. By discovering how jumping spiders walk, we can learn a lot about how behavioral characteristics can influence the evolution of mimicry traits in animals.
- Barber, J. R., & Conner, W. E. (2007). Acoustic mimicry in a predator–prey interaction. Proceedings of the National Academy of Sciences of the United States of America, 104(22), 9331–9334.
- Srygley, R. B., & Ellington, C. P. (1999). Discrimination of flying mimetic, passion-vine butterflies Heliconius. Proceedings of the Royal Society B: Biological Sciences, 266(1434), 2137.
- Golding, Y. C., & Edmunds, M. (2000). Behavioural mimicry of honeybees (Apis mellifera) by droneflies (Diptera: Syrphidae: Eristalis spp.). Proceedings of the Royal Society B: Biological Sciences, 267(1446), 903–909.
- Nelson, X.J. & Card, A. (2016). Locomotory mimicry in ant-like spiders, Behavioral Ecology, 27(3), 700-707.
Edited by Nikki Forrester