Imitation, flattery, and body-plans

If imitation is the highest form of flattery, then ants might be among the most celebrated creatures of the insect world. A delightful essay in Current Biology describes some of the more uncanny cases of ant-mimicry among arthropods, among them the treehopper Cyphonia clavata.

Ant-mimicking treehopper Cyphonia clavata. Via Flickr.

The 'ant' that's apparently riding on the back of the treehopper is actually an extension of its headshield. The green coloration of the rest of its body blends in with its leafy background, so on first glance you only see the black 'ant'. The mimic is seated in reverse: if you look carefully at the 'abdomen' of the 'ant', you'll see the green eye of the treehopper staring right back at you. This makes sense because in their defensive posture, ants move backwards.

Creatures that mimic ants are called myrmecomorphs ('ant-shaped', from Greek). There are certainly a lot of ants out there to be imitated (Antweb has high-resolution scans of ant specimens from around the world). One statement in this essay got me thinking:

"... there are about 2,000 species that mimic ants. Not surprisingly, these are nearly all insects or spiders, as a certain degree of body plan resemblance to ants is probably a prerequisite to becoming a myrmecomorph. ... ant mimicry must have evolved many times independently."

Treehoppers are true bugs (Hemiptera) in the family Membracidae. They are known for their headshields, also called helmets, which are responsible for the great diversity of form in this family. Some of them look like they may have inspired the fascinators worn by some of the women at the recent Royal wedding.

Diversity of treehopper headshield forms - Cyphonia clavata is at bottom right. Via Nature

Developmental biologists have now found that the helmet is actually a body plan innovation (original paper abstract), something that's incredibly rare in evolution. New structures are typically formed by modifying or reducing existing parts of anatomy, but innovations such as new appendages or new body segments are much rarer.

Insects typically have a pair of legs on each of the three thoracic segments (T1 to T3), and wings on the 2nd and 3rd (T2 and T3). The helmet arises from the 1st thoracic segment (T1), and anatomical observations have suggested in the past that they might be homologues of wings. By looking at gene expression in a developing treehopper, biologists found that wing-specific transcription factors (which control the expression of other genes), especially Nubbin, are expressed in the developing helmet, which indicates that it's developmentally homologous to wing appendages. They suggest that the Hox genes responsible for suppressing wing formation in T1 have been suppressed in treehoppers, allowing the evolution of the helmet. Because they aren't needed for flight, they aren't faced with the same physical constraints and so have been able to develop into a wild variety of shapes and sizes.

So coming back to the statement about body plan resemblances quoted above, perhaps the idea is more beguiling than it actually should be. An ant mimic for sure will have to be bilaterally symmetrical, of the right ant size, but beyond that I don't think the prerequisites for mimicry can be quantified. For conceptual proof of this, watch this famous octopus species mimic in turn a brittlestar, lionfish, and sea snake.