On each of the four club-like extensions (rhopalia) near the base of the cubozoan jellyfish bell there are two camera-eyes, one pointing upwards and the other downwards.

Cubozoans are the most extraordinary of the cnidarians, and in a number of respects are “honorary fish”; that is they show active locomotion, engage in courtship and copulation, hunt, are often highly venomous, sleep, and most famously have remarkably sophisticated camera eyes. On each of the four club-like extensions (rhopalia) near the base of the bell there are two camera eyes, one pointing upwards and the other downwards (although the rhopalia can turn and rotate).  Each is built on the standard pattern, with a spherical lens and the lower eye also possesses an iris which will contract when the intensity of light increases. It is a remarkable example of convergent evolution, especially with the vertebrates. The retina has a spectral sensitivity in the blue-green frequency, but to date there is no firm evidence for colour vision. In addition to the two camera eyes, each rhopalia possesses much simpler eyes in the form of a pair of pit eyes and a pair of slit eyes.

Eye function in cubozoans

So what exactly are the eyes used for? One puzzle is why each rhopalium has three different types of eye. Even more peculiar is that the lenses of the camera eye, and especially that of the upper eye, correct almost perfectly for spherical aberration so in principle the power of focussing should be excellent. So it is, but the problem is that the point of focus lies well behind the eye. Hence here is an animal with a world-class eye that produces a fuzzy image. Some cubozoans live amongst mangroves, and it is suggested the eye is adapted to navigate safely amongst the rocks and find the sunlit areas where their prey congregates. Tests on their ability to navigate certainly suggest an ability to move round objects, and in one experiment they successfully approached a piece of cord in the tank and then attached themselves (using an adhesive organ). It is also speculative precisely what function the pit and slit eyes play. The pit eyes are the simplest and may only be sensitive to light in the same way as other eye spots. The slit eyes, however, have a vitreous structure that presumably serves as a lens, and this lies above the photoreceptor layer. Although it is possible to plot out the fields of vision their actual role in vision is still undecided, but they may serve as adjuncts to the camera eyes.

Visual information processing

Another puzzle is that because cubozoans are cnidarians they have no real equivalent to a brain, let alone an optic lobe (or equivalent) so where is the visual signal processed?  It seems likely that at least some of the processing takes place in the nervous system of the rhopalia (and intriguingly it shows a bilaterally symmetrical organization), but some also takes place in the nerve ring that runs around the bell. In this way the pacemakers that determine the contractions of the cell operate and so control the direction of swimming. In sense the larvae of the cubozoans (the planula) go one better because as might be expected they have simple eyes (ocelli) but they appear to have no connection to the axons of the nervous system, and the response for the eye is obviously channelled using another mechanism, perhaps to adjacent microvilli that then “instruct” the locomotory cilia.

Developmental genetics of cubozoan eyes

Not surprisingly the molecular biology and developmental genetics have received some attention, and these too show interesting convergences. As might be expected the lens employs crystallins (Jl), but these have a quite different origin from other crystallins (which is the norm in this classic example of molecular convergence). Curiously whilst the occurrence of crystallins in the slit eye is not unexpected, they also are expressed in the pit eyes. Like other animals Pax-genes are employed in the development of the eye, but importantly it is not the canonical Pax-6 but the Pax 2/5/δ-type which evidently represents the predecessor gene. Equally interesting is that pathway for genetic instructions to build the components of the eye, notably the retina with its complement of opsins, is not like other invertebrates (as might be expected), but in fact is vertebrate-like. This is almost certainly convergent and it is intriguing to speculate why like vertebrates the cubozoans employ the phosphodiesterase cascade to construct a ciliary type of retina.