SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY NUMBER 89

SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY NUMBER 89 Avian Paleontology at the Close of the 20th Century: Proceedings of the 4th International Meeting of the Society of Avian Paleontology and Evolution, Washington, D.C., 4-7 June 1996 Storrs L. Olson EDITOR Peter Wellnhofer, Cecile Mourer-Chauvire, David W. Steadman, and Larry D. Martin ASSOCIATE EDITORS Smithsonian Institution Press Washington, D.C. 1999

Implications of the Cranial Morphology of Paleognaths for Avian Evolution Felix Y. Dzerzhinsky ABSTRACT In the early evolution of birds, bill formation produced a problem for muscular control of the thin, elongated upper jaw. In particular, it required a relatively high retracting force. Three sources of this force evolved. (1) A powerful M. retractor palatini (especially in Tinamiformes and Apteryx), originating primarily on the vomer and pterygoid, developed to provide direct muscular connection between the dermal palate and the cranial base. It apparently evolved due to a joining of the medial portions of the pterygoid and mandibular depressor muscles, which were aligned by development of the proc. mandibulae medialis (a character unique to birds). (2) The ancestral pseudotemporalis muscle developed into two portions, a large postorbital portion and an almost horizontally oriented intramandibular portion. Each portion seves to increase the retraction ability of the muscle as a whole. (3) The external mandibular adductor muscle developed, which, in neognaths, is larger than either muscle previously mentioned. Its evolutionary development was temporarily retarded by reduction of one of its places of origin the upper temporal arch. Introduction For more than a century, paleognaths have been subjected to morphological studies in order to ascertain their apparently primitive nature and to discover their position in avian phylogeny (W.K. Parker, 1866; T.J. Parker, 1891; Pycraft, 1900; Mc- Dowell, 1948; Hofer, 1945, 1950, 1955; de Beer, 1956; Webb, 1957; Muller, 1963; Bock, 1963; Cracraft, 1974; Yudin, 1970, 1978). I shall try to extract information on avian ancestry from the comparative and functional morphology of the feeding apparatus in paleognaths. Nomenclature for species' binomials and English names of modem birds follows Sibley and Monroe (1990). ACKNOWLEDGMENTS. In the process of this work I received valuable assistance from the following individuals and Felix Y. Dzerzhinsky, Faculty of Biology, Moscow State University, Moscow 119899, Russia. institutions. F. Vuilleumier and A.V. Andors (American Museum of Natural History, New York, New York) and E.G. Kordicova (Institute of Zoology, Kazakh Academy of Sciences, Almaty, Kazakhstan) granted me access to the alcoholic specimen of Apteryx sp.; K.A. Yudin and V.M. Loskot (Zoological Institute, Russian Academy of Sciences, St. Petersburg, Russia) provided me with specimens ofeudromia and Casuarius and with several skulls of paleognaths, respectively. M. V. Bevolskaya (Institute of Cattle-Breeding Askania-Nova, Ukraine) permitted me to dissect the alcoholic heads of Dromaius novaehollandiae (Latham), Rhea americana (Linnaeus), and Struthio camelus (Linnaeus). A. Elzanowski (National Museum of Natural History, Smithsonian Institution, Washington, D.C.) and E.N. Kurochkin kindly assisted me in obtaining literature. A.N. Kuznetsov helped me in editing the manuscript and in translating it into English. S.C. Bennett, L.D. Martin, and S.L. Olson read the English version with fruitful criticism and helped me in editing it. I am sincerely grateful to all these persons. This work was supported by The Cultural Initiative Foundation, Moscow, and by The Russian Foundation of Basic Researches (RFBR, grant N 96-04-50822). Skeleto-muscular Consequences of Bill Formation The adductory force of the mandible is transferred to the upper jaw through a food object. Resistance of the upper jaw to this force is produced (Figure 1) by a combination of tension on the ventral stalk (premaxillary and maxillary bones with palate caudally) and longitudinal compression of the dorsal stalk (frontal projection of premaxillary and premaxillary processes of the nasal bones). The longer the jaw grew, the greater the forces became, and, due to jaw lightening, the stresses became ever greater. The active forces necessary for normal grasping of food items must be supplied by muscles. The muscular force that creates tension in the palate and upper-bill floor also can accomplish ventral movement of the upper jaw by means of re- 267

268 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY compression FIGURE 1. Lateral view depicting the combination of forces in the upper bill of a bird (Struthio sp.) that produce resistance against the push from the mandible when grasping food. Lateral view. F'=useful force applied to object, R=rectractory (active muscular) force, N=push transferred to the braincase via dorsal upper-bill stalk. traction (backward shift) of the palate. Therefore, it is called the retracting or retractory force. The ancestors of birds apparently had no obvious source of such a force. Birds, however, have evolved the following three sources of retracting force. 1. M. retractor palatini (Figure 2) is the ventromedial part of the pterygoid muscle (of Moller, 1930, 1931), which is rather large in tinamous (Figure 2 A,B; Dzerzhinsky, 1983; Elzanowski, 1987), ostriches (Figure 2 D,E), and especially mapteryx (Figure 2 c). In paleognaths (sensu Pycraft, 1900) this muscle usually originates on the pterygoid and on the rear end of the vomer. Its caudal attachment is not situated at the midline on the base of the braincase, as in many neognaths (sensu Pycraft, 1900), but more laterally, near the caudal attachment of the occipito-mandibular ligament, i.e., on the medial part of ala tym- FlGURE 2. Retractor palatini muscle in ventral view: A,B, tinamou, Eudromia elegans; C, Kiwi, Apteryx sp. (scale=5 mm); D,E, Ostrich, Struthio camelus. aca=aponeurosis of insertion of M. pterygoideus caudalis; ad'=aponeurose of insertion of M. depressor mandibulae that is related to M. retractor palatini; art=aponeurosis of insertion of M. retractor palatini; Pt=pterygoid; Rtp=M. retractor palatini; Vom=vomer (A,C,D, superficial layer; B,E, deeper layers) (A,B, after Dzerzhinsky, 1983.)

NUMBER 89 269 c FIGURE 3. Interrelationship of parts of the pseudotemporalis muscle and epipterygoid or its apparent remainder (lig. epipterygoideum): A,B, Tuatara, Sphenodon punctatus (Gray); C,D, tinamou, Eudromia elegans. Ept=epipterygoid; Lep=ligamentum epipterygoideum; Ls=laterosphenoid; Ps=undivided M. pseudotemporalis; Psp=M. pseudotemporalis profundus; Pss=M. pseudotemporalis superficialis; Q=quadrate. (A,B, after Dzerzhinsky and Yudin, 1979; C,D, after Dzerzhinsky, 1983.) D panica. In adult Tinamiformes (e.g., Rhynchotus) these relations are obscured by later ossification, but they are quite clear in young Eudromia elegans Geoffrey Saint-Hilaire (Dzerzhinsky, 1983). Due to the occipito-mandibular ligament, the pterygoid muscle as a whole can act similarly to the retractor, but in contrast to it, via the mandible. 2. M. pseudotemporalis (part of the internal mandibular adductor) applies a retractory force to the mandible, and the force is transferred to the palate via the pterygoid muscle. In Sphenodon and lizards (Figure 3A,B), M. pseudotemporalis is undivided and originates mainly from the epipterygoid. In birds, the epipterygoid would limit the mobility of the quadrate, so it either has been replaced by a flexible ligament, as in tinamous (Figure 3 C,D; Dzerzhinsky, 1983), or has been completely reduced. Consequently, the pseudot-emporalis muscle has been divided into two portions, originating from two ends of the former epipterygoid. M. pseudotemporalis profundus originates on the tip of proc. orbitalis quadrati, and M. pseudotemporalis superficialis originates on the front wall of the braincase. M. pseudotemporalis profundus produces a retracting force rather effectively, irrespective of the particular direction of its fibers, because the resulting force is transferred to the braincase very caudally, through the quadrato-cranial joint (Figure 4). The main part of M. pseudotemporalis superficialis is well developed in paleognaths and occupies a considerable area of the temporal surface of the braincase (Figure 6A). But even here it passes rather FIGURE 4 (right). Contraction effect of the pseudotemporalis profundus muscle in the skull of the Common Raven, Corvus corax Linnaeus. Ap sp =immediate force; A=fmal force transferred to the braincase via the quadrate bone; Q=quadrate; Sq=squamosum.

270 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY steeply, i.e., at a great angle to the jugal bar, and so produces a small refractory effect. It may, however, include (e.g., in Rhea) a very inclined portion, the so-called intramandibularis muscle (Figure SB). In some birds another inclined portion of this muscle has evolved. The so-called caput absconditum (Hofer, 1950) apparently is a derivative of the main part of M. pseudotemporalis superficialis that is situated in the posterior temporal fossa of a typically diapsid ancestor (Figure 5A; Dzerzhinsky and Yudin, 1979). It is found in Sphenisciformes, Procellariiformes (Figure SB), and Pelecaniformes but never in recent paleognathous birds. 3. M. adductor mandibulae externus (Figure 6) acting on the upper jaw via the pterygoid muscle is a very important retractor in most birds (e.g., cranes). In paleognaths, however, it shows a rather modest development and, except Apteryx, does not spread up over the temporal wall of the braincase. Thus, its origin is limited to the zygomatic process of the squamosal. This restriction might have resulted from a reduction of the main ancestral origin of the muscle, the upper jugal arch. Otherwise, it might be a result of the change of functional requirements in the muscle during the course of development of the long bill and the cranial kinesis. I presume that the immediate ancestors of birds had an akinetic skull that possessed some prerequisites of cranial kinesis, such as a loose basipterygoid articulation (Yudin, 1970). It seems likely that kinetic mobility appeared first in the most loaded zone, i.e., within the slender upper jaw (Figure 1), and thus resulted in an archaic rhynchokinesis (Yudin, 1970, 1978). One of the questions about the functional morphology of the avian skull is the influence of sharp strokes, such as are associated with pecking or with accidental strokes against hard substrates while gathering grain or catching small, agile prey. In tinamous, the loose articulation of the frontal bone with the adjacent parietal and laterosphenoid (Figure 7) is equivalent to the so-called "articulating frontoparietal joint" described by Houde (1981) in early Tertiary North American carinates. It does not allow significant rotary movements of any cranial part, so in my opinion it is not associated with ancient mesokinetic mobility. Rather it is for damping shocks received along the dorsal bill stalk while pecking. The ventral stalk of the upper jaw was initially compliant, and it had to be supported by some solid framework able to transfer to the braincase large, but not dangerous, forces. This framework is formed by the bony palate, and among recent paleognaths it is strongest in Apteryx (Figure 8A), doubtless due to its specialization for probing. In tinamous, Rhea (Figure 8B), and, apparently, recent Casuariiformes, the main trajectory of compression stresses runs from the palatine process of the premaxillary bone to the vomer, then to the pterygoid, the quadrate, and finally via the quadrate's otic process to the braincase (Dzerzhinsky, 1983). In ostriches (Figure 8c), where the palatal processes of the premaxillary are missing and the vomer is partly reduced, com- Pss Pss A Psp aim aps FIGURE 5. Comparison of the pseudotemporalis muscle in lateral view: A, lizard, Cyclura nubilis Gray; B, procellariiform bird, Northern Fulmar, Fulmarus glacialis (Linnaeus) (mandible and side wall of braincase partly destroyed and removed). aca=aponeurotic insertion lobe of the caput absconditum of the pseudotemporalis superficialis muscle; Aex=M. adductor mandibulae externus; aim= aponeurotical lobe of origin of the M. intramandibularis; apm=aponeurosis of insertion of the pseudotemporalis profundus muscle; aps=aponeurosis of insertion of the pseudotemporalis superficialis muscle; Ca=caput absconditum of the pseudotemporalis superficialis muscle; Im=intramandibularis muscle, part of the pseudotemporalis superficialis muscle; Psp=pseudotemporalis profundus muscle; Pss=pseudotemporalis superficialis muscle; *=especially inclined part of the pseudotemporalis muscle. (A, after lordansky, 1990; B, after Dzerzhinsky and Yudin, 1979.)

NUMBER 89 271 Pss pression stresses run from the bill tip through the premaxillary and maxillary bones to the palatine and then almost directly to the apex of the basipterygoid process. In the roof of the mouth, the ostrich possesses a broad gap that is closed only by skin. This patch of skin gains a gliding support from the parasphenoidal rostrum via a long, thin, anterior extension of the vomer. Source of the Medial Mandibular Process B Pvl Dm FIGURE 6. Lateral surface of jaw adductors in lateral view: A, Rhea, Rhea americana; B, White-naped Crane, Cms vipio. Aex=M. adductor mandibulae externus; Ap=M. adductor mandibulae posterior; Dm=depressor mandibulae muscle; Pss=pseudotem-poralis superficialis muscle; Pvl=ventrolateral portion of the pterygoid muscle. (B, after Kuular and Dzerzhinsky, 1994.) The processus mandibulae medialis is highly specific for Aves. For example, in Gobipteryx, a fossil Mongolian bird, Elzanowski (1974) regarded this process as a distinctly avian character. The functional properties of the muscular portion (ventromedial portion of the pterygoid muscle) inserting on its tip are influenced significantly by the particular position of the tip. It is placed extremely high in the sagittal plane, so corresponding muscular forces pass almost through the pivot of the quadrato-mandibular joint (Figure 9A) and therefore apply a negligible adductory component to the mandible as compared to the retractory one. In the frontal plane, the tip of the process is extremely close to the midline, and therefore those muscular forces tend to rotate the caudal part of mandibular branch and so expand the lower jaw as a whole (Figure 9B; Yudin, 1961). The functional conditions discussed above, however, do not seem to account for the first steps in the evolution of the medial mandibular process. There is a peculiarity in the paleognath jaw musculature that is more useful in this respect: the abovementioned M. retractor palatini. I suggest that this muscle may have arisen by a joining of two muscular units the ventromedial portion of the pterygoid muscle and the depressor mandibulae muscle. Thus, their primitive interconnection via the caudal portion of the mandible formed a two-link chain that foreshadowed the recent M. retractor palatini (Figure 10). The crucial event in their further evolution has been an optimization of their ability to exert a single force, which has been ensured by alignment of both muscular links, due to the displacement Ls Fr FIGURE 7. Lateral view of tinamou skull (Tinamiformes), showing the loose articulation of the frontal with adjacent bones. Fr= frontal; Ls=laterosphenoid; Pa=parietal.

272 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY FIGURE 8. Comparison of palate structures in ventral view: A, Kiwi, Apteryx sp.; B, Rhea, Rhea americana; c, Ostrich, Struthio camelus. Probable paths for transferring longitudinal compression forces stippled. Bpt=basipterygoid process; Mx=maxillary; Pal=palatine; Pmx=premaxillary; Pt=pterygoid; Q=quadrate; Vom=vomer. Scale=20 mm. of their interconnecting point by means of the formation and gradual elongation of the medial mandibular process. After the alignment, this hypothetical digastric muscular complex must have separated from the mandible. Apparently, the recent occipito-mandibular ligament represents the reduced caudal belly of the digastric complex. Conclusion I would like to comment on the reinterpretation by McDowell (1978) of the homologies in the avian upper jaw and palate. It is, of course, tempting to use the kinetic mobility in the skull as a cause of fragmentation of a huge, ancient pterygoid bone into two; however, many traits in the general arrangement of the bones (primarily palatine position relative to the choana, premaxillary, etc.) seem to be consistent with the traditional interpretation that these two bones represent the reptilian palatine and pterygoid. The skull in ancient birds almost certainly had less internal mobility than it does in recent paleognaths, and such characters as the shape of the lateral rim of the palate or the pattern of epidermal papillae can hardly be valid. Finally, it is too difficult to accept McDowell's proposed loss of the maxillary bone in birds and his consequent thesis that the maxillopalatine of birds is equivalent to the reptilian palatine. Literature Cited Bock, W.J. 1963. The Cranial Evidence for Ratite Affinities. In Charles G. Sibley, editor, Proceedings of the XIII International Ornithological Congress, Ithaca 17-24 1962, 1:39-54. Bühler, P. 1985. On the Morphology of the Skull of Archaeopteryx. In M.K. Hecht, editor, The Beginnings of Birds: Proceedings of the International Archaeopteryx Conference, Eichstätt, 1984, pages 135-140. Eichstätt: Freunde des Jura-Museums, Eichstätt. Cracraft, J. 1974. Phylogeny and Evolution of the Ratite Birds. Ibis, 116:494-521. de Beer, G.R. 1956. The Evolution of Ratites. Bulletin of the British Museum (Natural History), Zoology, 4(2):59-70.

NUMBER 89 273 Dm Dm Pmm Pvm Pmm B Pmm FIGURE 9. Some functional effects of the medial mandibular process: A=skull of the Carrion Crow, Corvus corone cornix Linnaeus, in sagittal section, seen from left; B=skull of the Herring Gull, Lams argentatus Pontoppidan, with mandible broadened by contraction of the ventromedial portion of the pterygoid muscle, ventral view. F Pvm =force of the ventromedial portion of the pterygoid muscle; F Pvl+Pdl =force of its lateral portions; Pmm=processus mandibulae medialis; Pvm=ventromedial portion of the pterygoid muscle. (B, after Yudin, 1961). Dzerzhinsky, F.Ya. 1983. On the Feeding Apparatus ofeudromia elegans; On the Question of Morphological Specificity of the Feeding Apparatus in Paleognaths. Trudy Zoologicheskogo Instituta Akademii Nauk SSSR [Proceedings of the Zoological Institute of the Academy of Sciences of USSR], 116:63-118, 4 figures. [In Russian.] Dzerzhinsky, F.Ya., and K.A. Yudin 1979. On the Homology of Jaw Muscles in Tuatara and Birds. Ornitologiya (Moscow), 14:14-34, 7 figures. [In Russian; English edi- FIGURE 10. Hypothetical stages in the formation of the proc. mandibulae medialis in birds, drawn on the model of the skull of Archeopteryx (Buhler, 1985). Dm=depressor mandibulae muscle; Pmm=processus mandibulae medialis; Pvm=ventromedial portion of the pterygoid muscle; Rtp=M. retractor palatini. tion appeared in 1982 in Ornithological Studies in the USSR (Moscow), 2:408-436.] Elzanowski, A. 1974. Preliminary Note on the Palaeognathous Bird from the Upper Cretacious of Mongolia. Palaeontologia Polonica, 30(5): 103-1 09, 2 plates. [Results of the Polish-Mongolian Palaeontological Expeditions.] 1987. Cranial and Eyelid Muscles and Ligaments of the Tinamous (Aves:Tinamiformes). Zoologische Jahrbücher, Abteilung für die Anatomie and Ontogenie der Tiere, 116(1):63-118. Hofer, H. 1945.Untersuchungen über den Bau des Vogelschädels. Zoologische Jahrbücher, Abteilung für die Anatomie und Ontogenie der Tiere, 1950. Zur Morphologie der Kiefermuskulatur der Vogel. Zoologische Jahrbücher, Abteilung für die Anatomie und Ontogenie der Tiere, 70:427-556. 1955. Neuere Untersuchungen zur Kopfmorphologie der Vöögel. In A. Portmann and E. Sutler, editors, ActaXl Congressus Inlernationalis Ornithologici, Basel, 1954, pages 104-137. Basel: Birkhauser. Houde, P. 1981. Paleognathous Carinate Birds from the Early Tertiary of North

274 SMITHSONIAN CONTRIBUTIONS TO PALEOBIOLOGY America. Science, 214(4526): 1236-1237. Iordanski, N.N. 1990. Evolution of Complex Adaptations: Feeding Apparatus in Amphibians and Reptiles. 310 pages, 65 figures. Moscow: Nauca. [In Rus-sian.] Kuular, U.S., and F.Ya. Dzerzhinsky 1994. Trophical Adaptations of Cranes (Gruidae) as Seen from the Comparative and Functional Morphology of the Jaw Apparatus. 62 pages, 11 figures. Deposited Paper in VINITI, Russian Academy of Sciences, N 2904, Moscow State University, Moscow. [In Russian.] McDowell, S. 1948. The Bony Palate of Birds, Part 1: The Palaeognathae. Auk, 65(4): 520-549. 1978. Homology Mapping of the Primitive Archosaurian Reptile Palate on the Palate of Birds. Evolutionary Theory, 4:81-94. Moller, W. 1930. Über die Schnabel und Zungenmechanik blutenbesuchender Vögel, Teil I. Biologia Generalis, 6:651-726. 1931. Über die Schnabel und Zungenmechanik blutenbesuchender Vögel, Teil II. Biologia Generalis, 7:99-154. Muller, H.J. 1963. Die Morphologic und Entwicklung des Craniums von Rhea americana Linne, II: Viszeralskelett, Mittelohr und Osteocranium. Zeitschhft für die Wissenschafiliche Zoologie, 168( l/2):35-l 18. Parker, T.J. 1891. Observations on the Anatomy and Development ofapteryx. Philosophical Transactions of the Royal Society of London, series B, 182:25-134. Parker, W.K. 1866. On the Structure and Development of the Skull in the Ostrich Tribe. Philosophical Transactions of the Royal Society of London, 156:113-183, 15 plates. Pycraft, W.P. 1900. On the Morphology and Phylogeny of the Palaeognathae (Ratitae and Crypturi) and Neognathae (Carinatae). Transactions of The Zoological Society, London, 15(5, 6):149-290. Sibley, C.G., and B.L. Monroe 1990. Distribution and Taxonomy of Birds of the World. 1111 pages. New Haven: Yale University Press. Webb, M. 1957. The Ontogeny of the Cranial Bones, Cranial Peripheral and Cranial Parasympathetic Nerves, Together with a Study of the Visceral Muscles ofstruthio. Acta Zoologica (Stockholm), 38:81-203. Yudin, K.A. 1961. On the Mechanismus of Mandible in Charadriiformes, Procellariiformes, and Some Other Birds. Trudy Zoologicheskogo Institula Akademii Nauk SSSR [Proceedings of the Zoological Institute of the Academy of Sciences of USSR], 29:257-302, 30 figures. [In Russian.] 1970. Biological Significance and Evolution of the Cranial Kinetics in Birds. Trudy Zoologicheskogo Instituta Akademii Nauk SSSR [Proceedings of the Zoological Institute of the Academy of Sciences of USSR], 47:32-66. [In Russian.] 1978.Classical Morphological Characters and Recent Avian Systematics. Trudy Zoologicheskogo Instituta Akademii Nauk SSSR [Proceedings of the Zoological Institute of the Academy of Sciences of USSR], 76:3-8. [In Russian.] Zusi, R.L. 1984. A Functional and Evolutionary Analysis of Rhynchokinesis in Birds. Smithsonian Contributions to Zoology, 395: 40 pages, 20 figures.