As mentioned in the introduction, homology means that a character present in two or more lineages is derived from a common shared ancestor. In the case of the compound eye, this means that the common ancestor of both insects and crustaceans had a visual structure that could be classified as an early compound eye. Once this common ancestor split into the insect and crustacean lineages, both retained the compound eye, and its complexity increased through evolutionary time. 

The major argument supporting the homology of insect and crustacean compound eyes lies in their morphological similarities, including the structural components of the ommatidia. A concept, often termed Dollo's Law (1893), states that structures so complex and similar (like those of compound eyes in insects and crustaceans) should never evolve more than once. The similarities between the two groups seem so specific that there is no way they could have evolved independently or more than once. We see this conserved specificity of structure in insect and crustacean eyes, including the number and arrangement of cells in each ommatidium. Many insect and crustacean ommatidia contain 8 retinular and 4 crystaline cone cells per facet, and researchers believe this intricate detail is too specific to have evolved more than once (Oakley, 2003). Regarding the 8 retinular cells, 2 of those cells have rhabdomeres oriented in the mid-plane of the ommatidia, and one of these two cells directly transmits light wavelengths to the rhabdom. The other 6 retinular cells are oriented in pairs of 3 and are bilterally symmetric at the ommatidial axis. The 4 crystaline cone cells are also divided into two pairs, in which one pair are contiguous to each other at the ommatidial axis and the other pair is sector-shaped. Additionally, 4 interretinular fibers originating from the 4 crystaline cone cells stretch to the retinular cells, relaying light information. One of these fibers runs along the center axis, two show symmetry running perpendicular to the center axis, and the last fiber assymmetrical between the cone and retinular cells. This "pair" and "single" arrangement of cell orientation is seen in both insects and crustaceans, and the complexity of this cellular make-up would not seem likely to evolve more than once as accumulation of beneficial mutations is random and more than often deleterious. This conserved "pair" and "single" orientation of the ommatidial retinal and cone cells could be a synapomorphous trait, suggesting monophyletic origin among insects and crustaceans (Melzer, et. al., 1997).

However, some groups, both in insects and crustaceans, possess exceptions to the 8/4 rule, but this could be due to modifications favored by natural selection within lineages after the 8/4 compound eyes were homogeneously evolved. For example, how do we explain the reduction in cell numbers in certain groups of arthropods? One study identified a mutation in a gene of Drosophila melanogaster that resulted in a reduction of retinular cells from 8 to 7 cells (Harris, et. al., 1976). Theoretically, if this mutation was beneficial in a natural setting, the reduction of retinular cells from 8 to 7 would be favored by natural selection and likely persist in the population. Furthermore, how do we explain the absence or significant functional reduction of compound eyes in some groups of insects and crustaceans? Perhaps if the organism lives in solely dark environments, has a particuarly tiny body size, and does not have a large habitat range, there would be no need for an image-forming compound eye (Oakley, 2003). This is true for fleas and springtails (insects) and copepods (crustaceans). These few organisms are tiny, and springtails in particular live in low-light environments. Rather, these examples have ocelli (also present in all insects and crustaceans), which detect and directionalize light to aid in navigation. Compound eyes are not present in these groups because there is no need for their genome to expend energy on its development, and their small body sizes likely act as a developmental constraint. 

Ultimately, phylogenetic analysis of the distribution of ommatidial cell numbers is needed to further support the morphological arugment of homology (Oakley, 2003).

Additionally, similarities between insects and crustaceans have also been found at the molecular level, including a conserved monoclonal antibody, 3G6. Although the function of this antibody is not fully understood in arthropods, it has been found to be highly selective for expression in crystalline cone cells of the compound eye and evident in ancestral taxa like the bristle tail. It is emphasized that this highly conserved and specific molecular evidence is especially supported by the homology of the compound eye in insects and crustaceans (Edwards, et. al., 1990; Richter, 2002).

At length, it seems there is a great deal of evidence supporting the homology of compound eyes in insects and crustaceans, with most of the evidence being on a morphological basis. However, it is important to note that deriving morphological evidence for understanding phylogenetic relationships can lead to biases that could otherwise influence the results of sophisticated analyses. For example, subjective decisions must be made considering the relevance of similarities, the threshold to which variations can persist before they are considered non-homologous, and the likelihood that convergent evolution may explain the origin of the trait, such as functional or developmental constraints and selective pressures that would favor visual systems like the compound eye (Oakley, 2003). These subjective decisions limit our ability to fully understand how traits may have arose, as we cannot simply say that the 8/4 rule or the presence of ommatidia are the most relevant components for deciding homology among insect and crustacean compound eyes.

Thus, it is important that we consider both sides of the arugment - the non-homology of insect and crustacean compound eyes!

Image Sources