What is evolutionary convergence?

Blue-ringed octopus photoEvolutionary convergence occurs when unrelated organisms evolve similar adaptations to similar environmental or selective pressures, arriving there by very different routes. Here is a famous example. As you look out onto the world you do so through what we call a camera-eye, with a lens suspended between two fluid-filled chambers. Perhaps surprisingly, the octopus also has a remarkably similar camera-eye (as do squid and cuttlefish), yet we know that the octopus belongs to an invertebrate group called the cephalopod molluscs, evolutionarily very distant from the vertebrates (or speaking more widely the chordates), to which of course we belong. Our knowledge of animal relationships tells us that neither the common ancestor of molluscs nor chordates could possibly have possessed a camera-eye, so quite clearly they have evolved independently; the same solution has been arrived at by completely different routes.

The Map of Life documents many more examples of convergence for you to explore, and in fact in the case of the camera-eye alone there is much more to be said. The camera-eye has actually evolved at least seven times, most extraordinarily in a group of jellyfish known as the box-jellies (or cubozoans). Although these jellyfish have a nervous system, they don’t have a brain, and furthermore they belong to a phylum known as cnidarians, widely agreed to be amongst the most primitive of animals.

Convergence or secondary loss?

The presence of camera-eyes in one group of cnidarians may make you wonder whether they are therefore ‘primitive’ and it wouldn’t be simpler (or more ‘parsimonious’) to assume that they are present in octopus and vertebrates (and other examples such as snails and annelids) due to inheritance from very early in animal evolution rather than being convergent. Could the absence of camera-eyes in all the other groups then not be explained by secondary loss, whereby an ancestral trait is lost from its descendants?

Secondary loss is always possible, but molecular and ecological evidence points in exactly the opposite direction in the case of camera-eyes. Molecular phylogeny suggests that within the cnidarians, sea anemones and corals (a group known as anthozoans) are more primitive than box jellies, but they have no trace of an eye, let alone a camera-eye. In box jellies and the six or so other groups that have camera-eyes, we find a shared suite of characters, such as adaptations for hunting, a complex digestive system, specialised neurology and behaviour (for example uniquely among cnidarians, some species of box jelly engage in courtship followed by copulation). The animals concerned are also typically fast moving predators, as in the case of the active swimming cephalopods and box jellies. If we include this information, the distribution of animals that possess a camera-eye clearly makes much better sense in terms of convergent adaptation to a predatory mode of life rather than a common ancestry from the dawn of animal life.

Pax-6 and molecular convergence

The need to consider convergence from the level of molecules to whole organisms is exemplified by the genetics of eye development. Amazingly, it has been found that all animal eyes (from camera eyes to compound eyes) require the activity of a key gene called Pax-6 early in development in order to be specified. By genetic manipulation of embryos you can either provide excess Pax-6 to make eyes sprout all over the body (uck!), or you can transplant Pax-6 from a donor into a host animal, resulting in formation of a normal ‘host’ type eye. Does this mean that Pax-6 always “makes” camera eyes, so they are not convergent after all?

Box jellyfish photoSeveral points show that this objection is not valid. Although Pax-6 is indeed crucial for eye development (with a couple of exceptions), it does not work in isolation to build only one kind of eye. Pax-6 transcripts interact with a host of other genetic factors within taxon-specific developmental pathways, resulting in the formation of many kinds of eyes, from camera-eyes in octopus to the compound eyes that we see in insects. This is partly explained when we consider that the Pax-6 gene is very ancient, and almost certainly had a role in the early evolution of the brain area (supported by its expression in the brain and nose, as well as the eye of living animals). It is not surprising to observe that Pax-6 was recruited from its early role in brain specification to be a critical trigger for eye formation. Of note, in the box jellies a Pax gene is involved with eye formation but it is neither Pax-6 nor an immediate precursor, and also, some animals (such as nematodes) express Pax-6 but don’t develop eyes. So, Pax-6 is necessary, but not alone sufficient for eye formation.

Widespread convergence

Given that camera-eyes are constrained by having to work in a very precise way, with light entering the body and being precisely focussed on the retina, is it at all surprising that the camera-eyes of unrelated animals look the same? In one sense the repeated, convergent evolution of features such as camera-eyes is no big deal at all; organisms have to function in the real world, and it is no surprise to see that the same solutions to seeing, eating, moving and mating in particular environments have appeared again and again. Indeed, if there were only a few examples of convergence such as the camera-eye, we would necessarily conclude that when the call is out for something so specialized and seemingly “engineered”, then the constraints are so powerful that we would be more puzzled if we didn’t find convergence. However, the reality is that stunning examples of convergence can be found among all forms of life on this planet, and at all levels from molecules to societies. This ubiquity clearly illustrates that the forces of selection can lead unrelated organisms living in similar situations to acquire remarkably similar adaptive traits in response, resulting in repeated patterns that we are just beginning to explore.

Go to the top of the page