The Scleral Ossicles of Opisthocomus and Their Phylogenetic Significance

--Dried scleral rings of Opisthocomus, Galliformes, Cuculiformes, and other birds were examined to determine the pattern of ossicle overlap in Opisthocomus and its bearing on the phylogenetic relationships of this taxon. Although Opisthocomus shares a derived number of 12 scleral ossicles with cuculid Cuculiformes, the pattern of ossicle overlap differs. Nevertheless, fewer modifications are required to derive the number and pattern of ossicles in Opisthocomus from the conditions in cuculid or musophagid Cuculiformes than from the conditions in any galliform. Our findings also indicate that the scleral rings of birds, unlike those of lizards, often do not conserve overlap relations between adjacent ossicles during phylogeny. Received 2 March 1987, accepted 23 July 1987. THE phylogenetic relationships of the Hoatzin (Opisthocomus hoazin) have been the subject of a long-standing controversy in systematic ornithology. An enigmatic inhabitant of the flooded forests of Amazonia, Opisthocomus has been regarded as most closely related to many avian taxa, but most commonly to either Cuculiformes or Galliformes (Olson 1985; reviewed by Sibley and Ahlquist 1972, 1973). One character that has been used to support a close relationship between the Opisthocomus and Cucullformes is the morphology of the sclerotic ring, a ring of small bones that develops within the sclera in the corneal hemisphere of the eye in birds and many other vertebrates (Edinger 1929). Lemmrich (1931) found that the scleral ring of Opisthocomus consists of 12 ossicles per eye, which is lower than the 13-16 ossicles he found in 10 species of cracid and phasianid Galliformes (all of which showed a modal number of 14) but is identical to that found in a single cuculiform, Cuculus canorus. In addition to ossicle number, the pattern in which the scleral ossicles overlap varies among avian taxa (Lemmrich 1931, Curtis and Miller 1938). Because the same number of ossicles may have been derived independently, simple counts can be misleading. The pattern of ossicle overlap provides evidence about ossicle homologies and hence about cases in which the same number of ossicles has been derived through the loss of different individual ossicles. Although the pattern of scleral ossicle overlap is known for many birds, it has been unknown for Opisthocomus. We compared the numbers of ossicles and their patterns of overlap in the scleral rings of larger samples of Opisthocomus, Cuculiformes, Galliformes, and other relevant avian taxa. Our findings have implications both for the relationships of Opisthocomus and for the manner in which scleral