Because complex structural differences in adult extraocular muscles may have physiological and pathophysiological significance, the three-dimensional pattern of myosin heavy chain (MHC) isoform expression within the orbital and global layers of the muscle bellies compared with the distal tendon ends was quantitatively assessed. Three of the six extraocular muscles of adult rabbits were examined for immunohistologic expression of all fast, fast IIA/X, slow, neonatal and developmental MHC isoforms. The percentages of myofibers positive for each of these 5 myosin isoforms were determined in the orbital and global layers. There were relatively similar patterns of fast and slow MHC expression in the orbital and global layers of each of the three muscles examined. There were high levels of developmental MHC in the orbital layers, but significantly fewer developmental MHC positive myofibers in the global layer. The most variable expression was found with the neonatal MHC. There were significant differences between the longitudinal expression of the various isoforms in the middle of each muscle compared with the tendon end. In the orbital layer of all three muscles examined, the large numbers of fibers positive for fast MHC in the middle of the muscle dramatically decreased at the tendon end, with a concomitant increase in expression of slow myosin. There was a greater number of developmental MHC-positive myofibers at the tendon end than in the middle of the muscle in all three muscles examined. In the global layer, the IIA/X-positive myofibers comprised only half of the total number of fast-positive myofibers whereas in the orbital layer they comprised all or almost all of the fast positive myofibers. The configuration of the extraocular muscles is more complex than might be indicated by previous studies. The lateral rectus muscle had the most individual pattern of MHC expression when compared with the inferior rectus and inferior oblique muscles. Together with dramatic cross-sectional MHC fiber type differences between the orbital and global layers of the muscles, there are pronounced longitudinal differences in the proportions of myofibers expressing these five MHC isoforms in the middle region of the muscles and those in the distal tendon ends. This longitudinal progression appears to occur both within single myofibers, as well as within the series of myofibers that comprise the length of the muscle. We also confirm that the number of myofibers is reduced at the tendonous end while the cross-sectional area of each of the remaining myofibers is proportionally increased with regard to those in the muscle belly. Future studies may yet require two additional schemes for anatomic classification of the named extraocular muscles. One will be based on immunohistochemical features of their constituent myofibers as a supplement to classifications based on their electron microscopic appearance, innervation patterns or relative position with regard to the globe and orbit. Another will be based on the proportional length and longitudinal position of individual myofibers within an individual extraocular muscle.