In The Blink Of An Eye

by Andrew Parker

  Not surprisingly, the graded lens has been an evolutionary success in the water - it is also found in marine mammals, tadpoles, and some molluscs such as octopuses, squids, cuttlefishes, winkles, conches and pond snails. Michael Land, an authority on the eyes of living animals based at Sussex University in England, calculated that if nautilus had a camera-type eye of the same size as its pinhole eye, it would be four hundred times more sensitive and have one hundred times better resolution. Sensitivity refers to the ability of an eye to get enough light to the receptor cells, whereas resolution is the precision with which light rays from different directions are kept separate (to prevent blurring).

  Figure 7.3 Focusing of light rays (solid lines) by a graded lens (only three grades of material are shown). The dashed lines represent the paths induced by a standard lens of uniform material - the steeper angles of contact cause light to be bent more at the periphery. The core material of the graded lens, however, causes light to bend more than the material of the periphery layer, and so counteracts this angular discrepancy.

  In eyes of vertebrates on land, between 20 and 67 per cent of the focusing power is supplied by the cornea. So the lens can be designed specifically to correct blurring and to make accommodation for different distances - nearby objects would otherwise be imaged further away from the lens than distant ones. Mammals, birds and most reptiles meet these goals by changing the shape of their lens or cornea, thereby adjusting the position of the focal plane. They use tiny muscles to pull and stretch the lens. Alternatively, fish, frogs and snakes move their lenses backwards and forwards. Lens movement can provide adjustments for water or air in some amphibious animals.

  Although most spiders are equipped only with ocelli, jumping spiders and wolf spiders are exceptions. They have camera-type eyes with a thick cornea, which provides all the focusing power. The principal eyes of the jumping spider are unusual because, like birds of prey, they have a large pit in each retina that acts as a negative lens - it inverts and magnifies the image. This is comparable to the rear element of a telephoto lens in a camera.

  The principal eyes of the jumping spider are also unusual because their retinas are narrow vertical strips, giving a restricted visual field of only a few degrees in the horizontal direction but about 20° in the vertical. These retinas move sideways to compensate for their thinness, and so greatly extend the field of view in the manner of a photocopier scanning a picture. A similar mechanism is found in some snails and crustaceans. In some other crustaceans, such as the iridescent Sapphirina with the appearance of a swimming opal, there is a more extreme development. Here the retina is merely a dot with only a few light detection cells, but constant movement in all directions puts these cells to continual use.

  The iris controls the pupil size and, consequently, the amount of light entering the eye, in the manner of a camera diaphragm. But a further mechanism for brightness control may also exist in camera-type eyes, involving a reflector lying behind the retina. Like the scallop eye, the eyes of some vertebrates also contain a mirror behind their retinas, again employing the silver fish-mirror mechanism. But here the mirror functions so as not to focus the light - light is prefocused by a lens. In this case, the mirror provides an adaptation to the night. In the dark-adapted state, the reflector returns to the retina light rays that initially passed between the retinal cells without detection. So the most is made of the light available - anything not detected first time will have a second chance. The reflector in the eyes of cats and crocodiles reflects the beams of headlights and torches, often appearing obvious as ‘eyeshine’ at night. This represents the light missed both first and second time by the retina. When light levels are very low, all the rays striking the eye become invaluable to vision - the visual frontier between sight and blindness is approached. But when light levels are high, the reflector is redundant and becomes covered by a dark absorbing pigment. This mechanism is actually common in many nocturnal animals with camera-type eyes.

  Briefly stepping back over the visual threshold, into ‘not to see’ territory, some light perceivers actually contain a retina and a lens. Included among these are the receptors of scorpions, many web-building spiders and most snails. But their small size is important. This receptor type is not an ‘eye’ because it cannot focus an image on its retina - the retina is simply too close to the lens. The likely function of this receptor is to measure the average brightness or colour over large angles. As mentioned already, we are not so concerned with these light perceivers in this book, but they do help to illustrate just how much information can be acquired just from the architecture and size of a structure containing a lens and retina. The reason for writing about today’s eyes in this chapter is merely to provide a palaeontological tool. Where internal views of a simple eye-like structure are possible, we can calculate whether visual images were formed - whether it really was a simple eye. And the size information will become even more important when we consider fossils in relation to the Cambrian enigma. But fossil eyes that reveal internal architecture are extremely rare. So the best studied fossil eyes belong to the second type of eye found in animals, one where more information on sight can be deduced from just an outer surface. At least half the animals on Earth today are equipped with compound eyes.

  Compound eyes

  Figure 7.4 Scanning electron micrograph of the head of a fly, showing compound eyes.

  By way of introduction to compound eyes, I will return briefly to the somewhat unrefined ocelli. There is one group of bristle worms with ocelli but these are different from those of other animals. The difference lies in their arrangement. Lying on thick, feather-like filaments sprouting from the head, the ocelli of these worms are grouped together.

  Each ocellus has a sac-like region formed as an outgrowth of a sensory hair. This region lies within an infold in the skin of the animal and acts as a lens. Behind this lies a well-developed region of light-sensitive chemicals - the ‘retina’. And within a group of ocelli, light-absorbing pigment cells intermingle to prevent the same light rays affecting more than one ocellus. But the information collected by each ocellus is later combined strategically and so elaborate composite organs are formed. An organ of this type is known as a compound eye (although these particular eyes fall a little short of the visual mark).

  In contrast to the simple eye, the compound eye has multiple openings for light to enter - hence its name - and so always consists of numerous individual units, or ocelli, called ‘facets’. Other than minor appearances in the bristle worms and ark clams, the compound eye is a character of the arthropods. More precisely, compound eyes today occur in crustaceans, insects and horseshoe ‘crabs’ (which are actually more closely related to scorpions than true crabs). Compound eyes have evolved into sophisticated organs of sight, up to a third of the total body size in some seed-shrimps, and form images in different ways.

  The law of compound eyes was laid in 1891 in a monograph by the biologist Sigmund Exner, which became a landmark to both biologists and optical theorists. Exner broke all the rules of his day, where simple eye concepts were being applied to compound eyes. Instead Exner considered the focusing elements of compound eyes as ‘lens cylinders’. A conventional lens relies on the bending of light as it crosses a curved surface to focus rays. A lens cylinder, on the other hand, gently persuades light to change direction throughout the cylinder’s length. It is literally a cylinder, but one filled with graded material - graded in terms of its effect on light, just like the fish lens contemplated by Maxwell. The lens cylinder is most dense, and so causes light to travel slowest, along its central axis, its ability to bend light fading towards the edges. The overall effect of a lens cylinder is to provide the image-forming properties of a traditional lens. But there are alternatives to lens cylinders in some compound eyes.

  The compound eyes of many insects and crustaceans have a similar superficial appearance, but their focusing elements and mechanisms of image formation are very different. We can divide compound eyes into two b
asic types - apposition and superposition. The facets of apposition eyes are optically isolated from each other, so they each sample a different section of the environment. The tiny images formed within each facet are pieced together in the manner of a jigsaw puzzle to produce the complete picture. The facets of superposition eyes, on the other hand, cooperate optically so that they superimpose their light to form a single image at a common point on the retina. Dividing compound eyes further, there are variations of both apposition and superposition compound eyes in terms of focusing and image formation.

  Figure 7.5 Focusing mechanisms in the compound eyes of a) bees (apposition-type eye); b) moths and c) lobsters (superposition-type eyes). Graded material in b) and mirrors in c) (shown from the side and from above) achieve focusing. (Modified from Land, 1981.)

  Superposition eyes contain a large number of facets - up to several hundred. A broad zone of clear material separates the lenses from the retina below - the facets are not continuous tubes, so light can cross from the lens of one facet to the receptor of another. Exner discovered this after considering the cornea of the glow-worm (actually a beetle) as a single structure rather than as individual units. The cornea forms the lenses of the glow-worm, and Exner simply cleared out the interior of a glow-worm eye so that he could examine the complete array of lenses. Instead of a series of inverted images corresponding with each lens, he found just one upright image. So all the lenses must focus on to the same point on the retina.

  Some crustacean eyes are without lens cylinders, and it was not until 1975 that an alternative focusing mechanism was discovered. While examining the eyes of a crayfish and a deep-sea shrimp respectively, Michael Land and the German biologist Klaus Vogt independently found a superposition eye in which each facet was lined with mirrors. The mirrors were again similar to those found in fish skin, and formed mirror boxes, square in section. Exner got as far as illustrating the shape of these boxes, but overlooked the silver inner surfaces. It is now clear that if the boxes are considered with mirrored sides, the light rays will change direction as they are reflected and will all meet at the same point - on the retina where they form an image. In other words, mirror boxes perform the task of focusing.

  In 1988 a third form of superposition eye was discovered in many crabs by another contemporary expert on eyes - Dan-Eric Nilsson of Lund University in Sweden. The optics in this case are complex, and involve an intricate combination of ordinary lenses, cylindrical lenses, parabolic mirrors and light guides. The imaging mechanism is equally elaborate, with three separate systems in operation. Image formation can be predicted from the hardware alone, suggesting that fossil eyes could be informative after all.

  Compound eyes have no iris to control light levels, but they do have an alternative solution - they use dark pigment to remove some light where needed. This is similar to the way cats and crocodiles use pigment, but the pigment is in a different part of the eye. When superposition eyes are exposed to high light levels, the dark pigment moves between the lenses and retina to absorb a proportion of the light rays. When light levels become extreme, sometimes the facets become optically separated by dense collars or tubes of absorbing pigment so that the eye effectively acts as an apposition eye.

  All in all, the architecture or optics of modern eyes are well understood and can teach us much about the way their hosts see. And architecture can be preserved in the remains of extinct animals. We are now well enough informed to be able to browse the library on fossil eyes.

  Ancestral eyes

  The post-Cambrian view

  Conodonts are animals named after the Greek word meaning ‘cone teeth’. This is because for some time they were known only from jaw-like structures and bone fragments. Conodonts evolved in the Cambrian and became extinct 220 million years ago. They have been used extensively for comparing and assessing the age of rock sequences, but until the early 1980s we had no idea what the conodont animal looked like. Then complete fossil conodonts, around 340 million years old, were found in the Granton Shrimp Beds near Edinburgh, Scotland. These fossils revealed animals of eel-like appearance, with tails containing supporting fin rays . . . and heads with large, camera-type eyes.

  The smaller species of conodonts have eyes which are larger in relation to their body length than are the eyes of the larger species of conodonts in relation to theirs. This is consistent with a rule of the relative growth of the eye dating back to 1762, which holds that smaller animals have larger eyes in relation to their body size than do larger animals. From work on living vertebrates, eye size is known to influence visual sharpness. And the fact that conodonts possessed reasonably large eyes carries important information on early vertebrate evolution.

  Theories that conodonts are larval stages of ‘agnathans’, the primitive jawless fishes that include today’s lampreys and hagfishes, have not been well received. Agnathans are the first representatives of the true vertebrates, within the chordate phylum, which evolved around 485 million years ago, just after the Cambrian. But the relatively small eyes and large bodies of some conodonts are evidence against their interpretation as juvenile agnathans. So conodonts are more generally believed to be chordates close to, but not part of, the line leading to true vertebrates. The eyes tipped the balance of opinion in this direction.

  Among the living agnathans, only hagfishes have ‘eyes’ that are smaller than those of conodonts. But the ‘eyes’ of hagfishes, those primitive fishes caught in the deeper-water SEAS traps, are rather light perceivers - they do not yield visual images. They have probably degenerated as an adaptation to dark environments and burrowing. On the other hand, the eyes of lampreys are well developed and generally larger than those of conodonts. But there is one group of lampreys - the smallest brook lampreys - which offers clues to the vision of conodonts.

  Small brook lampreys have eyes that are about a millimetre and a half in diameter, equal in size to the eyes of the conodont Clydagnathus. There is evidence to suggest that similarity in the size of camera-type eyes reflects a similarity in cell and nervous complexity - the information processing system. The finding of eye muscles in another, well-preserved conodont, Promissum, supports this line of thinking - the muscles of similar-sized eyes are also equivalent.

  The conclusions drawn from a comparison with small brook lampreys are that conodonts had pattern vision and, as will become relevant later in this book, an active predatory lifestyle. Nevertheless, to say that smaller conodonts had relatively larger eyes does not imply that vision played a greater role in their behaviour. Instead they possessed visual organs near the minimum size limit for ‘eyes’ - these organs could not have produced visual images if they were smaller. Thorius, a miniaturised salamander which is the smallest land vertebrate living today, has camera-type eyes that are little over a millimetre in diameter, and this is believed to be the lowest size limit that will provide precise vision.

  At the larger practical limit of eye size was Ophthalmosaurus, a dolphin-shaped reptile between 3 and 4 metres in length. While dinosaurs were evolving big bodies on land, Ophthalmosaurus was setting a record for camera-type eyes in the sea. This animal really did possess eyes the size of soccer balls, and they were used to see at depths of 500 metres and more. It is thought that this reptile dived deep to avoid predators or to catch deep-dwelling prey. Unfortunately Ophthalmosaurus suffered from ‘the bends’, a condition familiar to deep-sea divers who approach the surface too rapidly. Rapid ascent causes nitrogen gas dissolved in the blood to decompress, forming bubbles that can block blood vessels and kill tissue. The bends leave visible depressions in the joints of bones, and those depressions are evident in the fossils of Ophthalmosaurus. The bends, and its effects, occurred much less commonly with its smaller-eyed ancestors, but the deep-diving, big-eyed version was stopped in its evolutionary tracks - and so the Ophthalmosaurus died out.

  Remaining within the ancient vertebrates, early fossil fishes commonly have dark stains in the region of the head. But what do these stains represent? The
eyes of one particular agnathan specimen can provide an answer. This fossil fish has, in the exact position of the stains found in other agnathans, an eyeball fully preserved in a sub-spherical hardened structure with a slit directed sideways. Alex Richie, an expert on primitive fishes at the Australian Museum, has interpreted the stains as the remains of pliable, presumably cartilaginous, structures surrounding the actual eyeball. The earliest of these camera-type eyes are known from a 430-million-year-old specimen of Jamoytius kerwoodi. Richie believes that although not specifically found in the eyes of this fossil specimen, there is no doubt that a lens of some form was present, based on comparisons with these camera-type eyes found in later, related species.

  So eyes have been found in fossils up to 430 million years old, and their vision has been extrapolated via comparisons with the eyes of today. But can we wind the clock back further and take a look into even older eyes?

  Back to the Burgess Shale

  After scrambling up a slippery section of the Canadian mountainside from the camp of Des Collins’s field team to the Burgess Shale quarries, I reached a ledge that had been excavated during both earlier and current fossil expeditions. At the back of the ledge was the exposed face of the quarry, where the layers of sediment were clearly defined through their various colours. On the ledge itself was a wooden table, which supported the fossils unearthed during the current excavations.

  Standing in the viewing area of the Burgess quarry, on the edge of Emerald Lake below, all that could be seen of the site above us was an indiscernible blue object. Along with the many tourists using the telescope provided, I wondered what it could be. It was, as I discovered on reaching the quarry, just an old plastic sheet, but one with an important purpose - to protect the latest fossil treasures laid out on the table from the harsh climate. These fossils were awaiting quality control, to ascertain their future in the display cases of museums all over the world. And there really were some treasures. The fossils whose photographs I had seen in coffee-table books and on the projector screens at a number of famous lectures were there in front of my very eyes. And I was one of the first people to see these specimens - they were fresh from the rock. But they were very well preserved and defined - and I could identify them all.