The optical mechanism of the eye of Limulus

Abstract

THE compound eyes of many marine and some terrestrial arthropods have smooth surfaces with no convex curvature to the facets. This means that image formation by ordinary spherical refraction is impossible. Nevertheless, apposition eyes of this kind form images behind each surface facet^1,2 (Fig. 1b) and the question has repeatedly arisen as to how these images are produced. Exner chose the king crab, Limulus, as providing the definitive example of such an eye, and from a mixture of observation and theory came up with the idea of a lens-cylinder, a flat-ended optical device in which the refractive index decreases from the central axis to the periphery in a roughly parabolic manner¹. This arrangement will form an image (Fig. 2a), and devices of this kind have actually been manufactured^3,4. Exner's theory has long been considered to be the correct account of image formation in this type of eye⁵. However, Levi-Setti, Park and Winston² have recently produced a fundamentally different explanation of image formation in Limulus eyes (Fig. 2b). This was based on the idea that each optical element, or crystalline cone, concentrated light by reflection, not refraction, and that it was the shape of the crystalline cone's reflective surface, and not its internal refractive index gradient, that led to the image-forming properties of the structure. This principle had already been used in another optical device, the ‘ideal light collector’ (ref. 6) which was invented as a tool for concentrating faint radiation. In this report I discuss the two suggested optical mechanisms for the eye of Limulus, and present evidence from interference microscopy that Exner's theory is probably the correct one.