Meeting Abstract
Eye structure directly limits what an animal can see. By studying the natural range of eye designs we can learn a great deal about visual ecology, visual development, and the selective pressures placed on eye development that drive the evolution of vision. The compound eye found among arthropods contains an impressive range of designs and sizes and is a prime subject in the study of vision. Compound eyes are composed of many units called ommatidia, each equipped with a lens that focuses light upon photoreceptors below. In contrast to camera-type eyes, many of the structures limiting compound eye performance are externally visible. For spherical eyes, many visual parameters like spatial resolution, optical sensitivity, and field of view can be measured externally using microscope images. For non-spherical eyes, which often have skewed ommatidial axes, internal structures must be accounted for. MicroCT, a burgeoning imaging technique, generates a 3d image of the specimen where voxel values represent physical densities. In particular, MicroCT can reliably image insect brain structures, like the visual neuropils, and eye structures, like the the crystalline cones of the ommatidia. We propose two methods, one for microscope and another for microCT images, and offer an open source Python program to semi-automatically approximate parameters like spatial acuity, optical sensitivity, and field of view across different regions of the eye. We demonstrate the reliability of these methods on the eyes of a number of insect species (fruit fly, moth, bee, ant), finding that both succeed in characterizing the optical performance of compound eyes accurately and reliably while minimizing labor.
