Early vertebrate eyes: a fossil record with limits

The evolution of eyes is often illustrated as a smooth climb from a light-sensitive patch to a camera-like organ. That diagram can help explain possible intermediate functions, but it is not a fossil sequence in which every stage is preserved. Early vertebrates were soft-bodied, and eyes are made largely of tissues that usually decay. Most evidence about their origins therefore comes from a combination of fossils, living animals, development and genetics.

Why eye fossils are unusual

The lens, retina and nerves are soft tissues. They are unlikely to survive burial and geological alteration. A fossil may preserve the outline of a skull but lose the tissues that would show exactly how an eye worked. In rare deposits, chemical traces or pigment-bearing structures can survive and be detected with specialised imaging. Such exceptional fossils are valuable precisely because they are uncommon.

That scarcity does not make the evolutionary history unknowable. Researchers compare the anatomy of living vertebrates, the development of eyes in embryos and the genes that build photoreceptors and retinal tissues. These sources can indicate which components are ancient and how different parts were modified. They answer questions that a fossil alone cannot, while fossils constrain the timing and anatomical context of particular lineages.

What the Silurian study adds

In 2026, researchers examined specimens of Jamoytius kerwoodi and Lasanius, two enigmatic jawless vertebrates from Silurian rocks in Scotland. Synchrotron radiation allowed the team to map elements within compressed fossils at a scale that is difficult to achieve with conventional light. The study reported information about the eye structures of Jamoytius as well as mineralised tissues in both forms.

The result matters because these animals sit among early vertebrate branches for which the fossil record is fragmentary. A preserved eye-region signal can inform hypotheses about how early vertebrates organised vision. It does not show that every fish at that time had the same eye, nor does it establish a straight line from Jamoytius to modern species. Fossils provide branch-specific snapshots, not a complete family movie.

New imaging can also change what scientists notice in specimens that museums have held for decades. The bones and impressions did not suddenly form in 2026; the analytical method made previously hidden chemical contrasts measurable. This is a different kind of discovery from finding a new fossil bed, and the distinction helps explain why old collections remain scientifically useful.

From light detection to image formation

Light-sensitive cells can detect brightness without forming a detailed image. A shallow cup can provide directional information; a pinhole-like opening can sharpen that signal; and a lens can focus light onto a retina. These arrangements offer a way to think about possible functions, but living animals with different eye designs are not necessarily preserved ancestors in sequence. Similar features can evolve independently, and an organism's visual performance depends on neural processing as well as eye shape.

Vertebrate eyes also share molecular and developmental components. Opsins respond to light, photoreceptors convert it into electrical signals, and a developing retina forms connections with the brain. Comparative work across vertebrates and their relatives helps identify which molecular systems are old and which structures were reorganised later. Those inferences are supported by biology across living species, not by soft-tissue fossils alone.

What we should not claim

The fossil record does not show a single universal progression that every animal followed. It cannot tell us the exact images an extinct fish perceived, and one preserved eye cannot represent all early vertebrates. Nor does a complex fossil eye disprove gradual evolution: complexity can accumulate through changes that have useful effects at each stage, while the sequence of those changes may be incompletely sampled in rocks.

Instead, paleontologists combine preserved structures, chemical signatures, comparison with living eyes and developmental evidence. Each source has a different strength. A fossil constrains what existed in a particular animal; genes and embryos reveal shared developmental machinery; anatomy and behaviour help test what a structure could do. The picture becomes stronger when independent evidence lines agree and uncertainty remains explicit when they do not.

The 2026 Jamoytius analysis illustrates that approach. It adds a new, measurable signal from rare soft-bodied fossils, while leaving much of the earliest history unresolved. For a related example of how preserved chemistry can inform anatomy without restoring an entire animal, see the evidence for colour in dinosaur feathers and skin.