Thomas Koppe / Hiroshi Nagai / Kurt W. Alt

The Paranasal Sinuses of Higher Primates Development, Function, and Evolution

Special Reprint

The Phylogenetic History of Paranasal Air Sinuses Lawrence M. Witmer

Chapter 2 The Phylogenetic History of Paranasal Air Sinuses

Lawrence M. Witmer

Paranasal air sinuses have long been fossil record is often critical in bridging of great interest to anatomists working morphologic gaps observed in the extant on mammals in general and primates realm, and a phylogenetic perspective in particular; clinically oriented human dictates which are the most appropriate anatomists probably have generated the paleontologic comparisons. largest proportion of the literature. Paranasal pneumaticity, however, is not restricted to mammals, and numerous other clades of vertebrates exhibit air-filled Definitions epithelial diverticula of the nasal cavity. Although paranasal sinuses in nonmam­ Before the distribution of pneumatic struc­ malian vertebrates often have been given tures throughout vertebrates can be the names of their presumed mammalian examined, a few concepts must be clari­ counterparts, the homologies of some of fied. The first is the distinction between a these sinuses have only recently been soft tissue pneumatic diverticulum and subjected to scrutiny (Witmer, 1995b). the bony recess that often houses it. The Progress in interpreting the diversity of term sinus has been applied to both, and, vertebrate sinuses historically has been for many questions, the distinction is not hampered by typologie approaches that that important. However, it often is crucial have sought to homologize mammalian in comparisons across broad taxonomic (principally human) sinuses with those of groups. For example, an epithelial diver­ other vertebrates under the often unstated ticulum may be present in numerous taxa, belief that all organisms will display mani­ but only pneumatize bone in a subset of festations of the archetype. A phyloge­ these taxa, an example being the subor­ netic perspective, however, improves bital diverticulum of the antorbital sinus in prospects for successful interpretation of birds and some other archosaurs (Witmer, sinus diversity, in that it provides for a bet­ 1997). Likewise, very similar bony recess­ ter sense of what is truly homologous, es may result from pneumatization via dif­ which in turn provides the appropriate ferent pneumatic diverticula (Witmer, framework for functional and other infer­ 1990), thus presenting problems of ences. Furthermore, evidence from the homology. Perhaps most fundamentally,

21 The Phylogenetic History of Paranasal Air Sinuses diverticula and bony recesses stand in nasal cavity proper. However, in those morphogenetic relation to each other as nonmammalian vertebrates eXhibiting cause and effect; that is, in most if not all paranasal pneumaticity, diverticula arise cases, the bony recesses are actually pro-­ not only from the nasal cavity proper, but duced through pneumatization via the also from the other portions of the cavitas epithelial diverticulum and its osteoclastic nasalis. Thus, although mammals gener­ front (Witmer, 1995a, 1997). ally have extensive paranasal sinuses, the Another distinction must be made morphogenetic diversity of sinuses in between intracapsular and extracapsular mammals is less than in some other ver­ recesses. Intracapsular recesses are tebrate clades. those internal to the cartilaginous nasal capsule, and, although their names some­ times sound like pneumatic recesses (eg, Mihalkovics' [1898] identification of a Phylogenetic Distribution of "sinus maxillaris" in the lizard Lacerta), it Paranasal Sinuses appears that only epithelial diverticula that have escaped the bounds of the cartilagi­ The literature on the phylogenetic distrib­ nous capsule (that is, extracapsular sinus­ ution of sinuses is widely scattered es) are competent to pneumatize bone among the primary literature. Summary (Witmer, 1995b). In many clades of verte­ information can be found in comparative brates, the nasal capsule is not a com­ anatomic texts, especially older ones, plete cartilaginous box, but rather has var­ such as Dieulafe (1905a, 1905b, 1905c, ious openings (in addition to the nares 1905d), Plate (1924), and Matthes (1934), and choanae) that may transmit evagina­ among others. Nemours (1930a, 1930b) tions of nasal epithelium, allowing, in published two rather naive and very typo­ effect, intracapsular epithelium to logic reviews on the status of the become extracapsular. As discussed later paranasal sinuses of modern amphibians, in this chapter, some of these outpocket­ reptiles, and birds. Negus (1958) pub­ ings bear some histologic resemblance to lished .an extensive comparative treatise pneumatic diverticula, but only a few that provides excellent information on actually pneumatize bone. mammals but only superficial coverage of A final point of clarification relates to the nonmammalian vertebrates. The closest site of origination within the nasal cavity of thing to a modern review is Witmer's the pneumatic diverticulum. Parsons' (1995b) study of facial anatomy of (1970) division of the nasal cavity is useful archosaurs, which reviews much of the in this regard. He divided the nasal cavity previous literature on other extant into the vestibule rostrally, the nasal cavity amniotes. This chapter is intended only as proper (cavum nasi proprium) in the mid­ an overview, and there are a number of dle, and the nasopharyngeal duct caudal­ areas that require much more in-depth ly. In primates, as in apparently all mam­ study. In the following survey, vertebrate mals (Paulli, 1900a, 1900b, 1900c; taxa will be taken in turn, roughly following Moore, 1981), all of the various paranasal the pectinate cladogram topology of Fig sinuses originate as diverticula of the 2-1.

22 Phylogenetic Distribution of Paranasal Sinuses

of why some paranasal diverticula pneu­ matize bone and others do not. Given this situation, it would be of interest to survey fossil material of the large-skulled "reptil­ iomorph" nonamniote anthracosaurs to seek evidence of paranasal sinuses.

Amniota

Despite any prospects for the discovery of paranasal air sinuses in nonamniote ver­ tebrates, it remains that paranasal pneu­ maticity is still known only in amniotes. Tetrapoda However, as discussed in detail elsewhere Fig 2-1 Phylogenetic relationships of extant (Witmer, 1995b), it appears that the pres­ tetrapods. (Topology after Gauthier et aI., 1988.) ence of paranasal sinuses is not a synapo­ morphy of Amniota. Rather, pneumaticity evolved independently at least twice-in Mammalia and Archosauria-and, in each case, numerous different paranasal sinus­ es evolved. The following will be a very Nonamniote vertebrates brief survey of the major clades of amni6tes (see Fig 2-1), examining their To my knowledge, bony pneumatic various epithelial diverticula and bony recesses have not been reported in any pneumatic recesses, when present. extinct or extant nonamniote vertebrate (i.e., fishes and amphibians), a finding that Mammalia is perhaps not too surprising, given the aquatic habits of most of these animals. Paulli's (1900a, 1900b, 1900c) masterful However, the potential for pneumatization work on mammalian paranasal pneu­ of bone would seem to be present in maticity remains the primary reference some modern amphibians, in that extra­ and was the basis for most of the recent capsular epithelial diverticula have been reviews (Moore, 1981; Novacek, 1993), described and/or illustrated for various although Negus (1958) also provided salamanders and especially frogs by some novel Observations. According to Mihalkovics (1898), Plate (1924), Ne­ these sources, monotremes and marsupi­ mours (1930a), Matthes (1934), and als (with perhaps a single exception) lack Jurgens (1971), among others. paranasal sinuses of any kind. Fur­ These diverticula appear in section (Fig thermore, among eutherians (i.e., placen­ 2-2A), to be generally similar to the diver­ tal mammals), only the maxillary sinus is a ticula that pneumatize bone in amniotes, nearly ubiquitous feature. (It is absent in a but they do not do so, raising the question few small bats [Novacek, 1993] and most

23 The Phylogenetic History of Paranasal Air Sinuses

A nasal capsule B nasal cavity vestibule proper

bone nasal cavity proper ....-;:;;;/~'l'/~;;!';

o bone C recessus ducti nasopharyngei

nasal nasal cavity nasal nasal cavity capsule proper capsule proper

Fig 2·2 Extracapsular epithelial diverticula without pneumatization of bone in (A) the anuran amphibian, Hy/a arborea (transverse section); (8) the turtle, Testudo graeca (sagittal section); (e) the turtle, Emys eumpaea (transverse section); and (D) the squamate lepidosaur, Varanus ni/oticus (transverse section). Pneumatization of bone requires the evagination of a thin-walled epithelial diverticulum beyond the bounds of the cartilaginous nasal capsule. The taxa shown have such a diverticulum, but it fails to pneumatize bone. In Varanus, the diverticulum has been named the recessus extracapsu/aris (Malan, 1946), whereas in turtles it is known as the recessus ducti nasopharyngei. (A) Modified from Mihalkovics (1898). (8) Modified from Parsons (1970) and Witmer (1995b). (C) Modified from Mihalkovics (1898). (D) Modified from Malan (1946).

marine mammals [Moore, 1981]). It ap­ from the olfactory region of the nasal cav­ pears that all of the eutherian maxillary ity. This system may be best termed the sinuses are homologous, although a rig­ ethmoidal sinus system, although the orous test of this hypothesis has not been terms frontal and sphenoidal have also published. been applied. Sinuses derived from the Whereas the maxillary sinus derives ethmoidal system have evolved in a num­ from the respiratory portion of the nasal ber of eutherian clades, and the homolo­ cavity, some eutherians have another sys­ gies of many of the sinuses are far from tem of sinuses that derives more or less clear; it seems certain that many of the

24 Phylogenetic Distribution of Paranasal Sinuses described "frontal" and "sphenoidal" horned turtle Meiolania (Gaffney, 1983), sinuses have appeared independently. In which has an intraosseous cavity associ­ humans, it is this ethmoidal system that ated with the naris; should this prove to be produces the frontal and sphenoidal truly pneumatic, it would likely represent a sinuses, as well as the definitive ethmoidal diverticulum of the nasal vestibule rather air cells. than of the nasal cavity proper. Although the presence of a maxillary sinus would seem to be an apomorphy of Lepidosauria Eutheria (given its absence in marsupials Lepidosaurs (i.e., lizards, snakes, amphis­ and monotremes), this assessment may baenians, and Sphenodon) can be be premature, in that a number of non­ regarded as lacking any paranasal pneu­ mammalian cynodonts (i.e., mammal-like maticity. Certainly no evidence of bony reptiles) have been described as possess­ recesses has been reported. In some ing maxillary sinuses (Witmer, 1995b). lizards, an intracapsular space, known as Thus, although the skulls of many modern the extraconchal recess, becomes mammals are extensively pneumatized by exposed laterally by a gap in the wall of the nasal cavity, there is still a great deal the cartilaginous capsule. This recess about the pattern of evolution of these bears some resemblance to a paranasal sinuses that remains to be elucidated. diverticulum, again leading Mihalkovics (1898) and Nemours (1930b) to regard some lizards as possessing "maxillary Testudines sinuses." In Varanus, the extraconchal recess is With the possible exception of a single especially extensive (see Fig 2-20), taking fossil taxon, no turtles have been a fully extracapsular course to lie adjacent described as having any bony pneumatic to the bones of the snout (Malan, 1946); recesses associated with the nasal cavity. although this recess never pneumatizes However, turtles do possess an extracap­ these bones, this is probably the nearest sular epithelial diverticulum, arising not approach to a paranasal sinus among lep­ from the nasal cavity proper, but rather idosaurs. from the nasopharyngeal duct (Parsons, 1970). This recessus ducti nasopharyngei Archosauria evaginates laterally from the rostralmost end of the nasopharyngeal duct, but does Archosaurs (crocodilians, birds, nonavian not pneumatize bone in any known taxa dinosaurs, pterosaurs, etc.) present prob­ (see Fig 2-28, C; Witmer, 1995b). Mihal­ ably the greatest known diversity of kovics (1898) and Nemours (1930b) paranasal air sinuses and bony pneumat­ referred to the recessus as the "maxillary ic structures (Witmer, 1997), which is not sinus," but it clearly is not homologous too surprising, given that Archosauria is to the mammalian structure (Witmer, today the most specios group of terres­ 1995b). trial vertebrates and has been so for prob­ The only candidate for a bony pneu­ ably the past 200 million years. The pneu­ matic recess in turtles is the Pleistocene matic diverticula of archosaurs are best

25 The Phylogenetic History of Paranasal Air Sinuses

subsidiary diverticula within -.....z:::.~~;;:?\l suborbital accessory cavities r- diverticulum

antorbital SinUS suborbital diverticulum

Fig 2-3 Antorbital paranasal air sinus of archosaurs in a goose, Anser anser (left), and a nonavian thero­ pod dinosaur, Allosaurus fragilis (right). The archosaurian antorbital sinus is often only partially enclosed in bone (as shown) and is almost always covered only with skin laterally. The antorbital sinus often has a num­ ber of subsidiary diverticula that may pneumatize surrounding bones, resulting in accessory cavities (shown in Allosaurus). One such subsidiary diverticulum, the suborbital diverticulum, interleaves with jaw musculature, providing a mechanism for actively ventilating the paranasal sinus system. (Modified from Witmer, 1997.)

discussed on the basis of their site of variety of subsidiary diverticula that pneu­ origin, in the nasal cavity proper, the matize bone, producing so-called acces­ nasopharyngeal duct, or the vestibule. sory cavities within the surrounding bones. These subsidiary diverticula are best developed in the theropod dinosaurs Nasal cavity proper (including birds), producing accessory Based on studies of extant and extinct cavities in the maxillary, lacrimal, nasal, archosaurs (Witmer, 1995b, 1997), a sin­ jugal, and/or palatine bones. In some gle epithelial diverticulum of the nasal cav­ theropods, the entire facial skeleton is ity proper can be inferred to be present in intensely pneumatized by these diverticu­ all archosaurs. This air sac, known as the la, reducing the facial skeleton to a series antorbital sinus (equivalent to the cavicon­ of struts between large, air-filled spaces. chal sinus of crocodilians), is lodged in a Subsidiary diverticula of the antorbital bony cavity (the antorbital cavity) formed sinus are not limited to theropods, and primarily by the maxillary and lacrimal accessory cavities are found also in vari­ bone, with varying contributions from the ous taxa of prosauropod and ornithischi­ nasal and/or jugal bones (Fig 2-3). an dinosaurs, pterosaurs, and a few basal The antorbital sinus itself often has a archosaurs (Witmer, 1997).

26 Phylogenetic Distribution of Paranasal Sinuses

postvestibular ~~:--vestibule recess nasal cavity proper

caudo­ lateral recess

prefrontal recess /'....•..;.'.'.•. 1,". '.'./ I,

Fig 2·4 Snout of Alligator mississippiensis in dorsal view (horizontally sectioned with roof removed), Parts of the nasal cavity are labeled at right; the broken arrow shows the course of the nasopharyngeal duct. Paranasal sinuses that derive from the nasal cavity proper are labeled at left (solid arrows), (Modified from Witmer, 1995b,)

A remarkable difference between the one of the subsidiary diverticula, the sub­ antorbital sinus of archosaurs and the orbital diverticulum, extends caudally into paranasal sinuses of mammals is that, the orbit beneath the eyeball to interleave whereas the mammalian sinuses are gen­ with the adductor musculature (see Fig 2­ erally fully enclosed in bone (except for 3). Thus, as the mandible is abducted and the ostium), the archosaurian antorbital adducted, negative and positive pres­ sinus is in most cases only partially sures, respectively, are set up within the enclosed in bone and is covered with skin suborbital diverticulum and hence the laterally (see Fig 2-3). Furthermore, in whole antorbital sinus system. These some taxa, such as birds and probably at pressure changes allow the sinus system least some nonavian theropod dinosaurs, to be actively ventilated, with mandibular

27 The Phylogenetic History of Paranasal Air Sinuses

A

nasal cavity proper secondary choana nasopharyngeal B duct

pterygoid bulra

Fig 2-5 Pterygoid bulla of the crocodilian, Gavialis gangeticus, in (A) medial view (sagittally sectioned skull) and (8) ventral view. The pterygoid bulla is an inflated paranasal air sinus derived from the nasopharyngeal duct. (Modified from Kalin, 1955.)

Fig 2·6 Narial crest of the lambeosaurine hadrosaurid dinosaur, Corythosaurus casuarius. (A) Left lateral view of skull. (8) Left lateral view of passages and chambers within the narial crest. (C) Rostral view of skull with narial crest transversely sectioned to reveal cavities. The nasal vestibule is greatly hypertrophied, as are the bones that surround it (the premaxilla and nasal bones), and the whole apparatus is mounted above the rest of the skull. The lateral diverticula may be viewed as examples of paranasal sinuses derived from the nasal vestibule. (Modified from Weishampel, 1981.)

28 Phylogenetic Distribution of Paranasal Sinuses

A nasal

B c central chamber

lateral diverticulum

29 The Phylogenetic History of Paranasal Air Sinuses movements acting as a bellows pump as the vomerine bulla of alligatorines, the (Witmer, 1997). pterygopalatine bulla within the orbit of Such a powered ventilation system is many crocodilians, and, most dramatical­ unique in vertebrates. Mammalian sinus­ ly, the enormous and bulbous pterygoid es are dead-air spaces, with air exchange bulla of the gharial, Gavialis gangeticus occurring only very slowly through diffu­ (Fig 2-5). Finally, the pterygoid sinus of alli­ sion. Interestingly, modern crocodilians gators is a remarkable, invasive pneumat­ have lost the open construction of the ic recess with many communicating antorbital cavity and lack a suborbital chambers that extends dorsally from the diverticulum (Fig 2-4) (Witmer, 1995b, nasopharyngeal duct in the vicinity of the 1997); as a result, they have converged secondary choana. with the mammalian system and lack active ventilation. Vestibule The antorbital sinus system is clearly the most important of the pneumatic sys­ The third and final division of the nasal tems of archosaurs because it is found in cavity, the vestibule, also is a source of all known archosaurian taxa and truly pneumatic diverticula and bony recesses dominates the facial morphology of in a few archosaur taxa. Best known are some of them. For example, in some the cavities within the narial crests of lam­ archosaurs, the bony antorbital cavity and beosaurine hadrosaurid dinosaurs (Weis­ its accessory cavities exceed the com­ hampel, 1981; Weishampel and Horner, bined volumes of the brain and all cephal­ 1990), In lambeosaurines, the nasal ic sense organs. The antorbital sinus, vestibule is greatly enlarged, and the pre­ however, is not the only sinus system maxillary and nasal bones are expanded known in archosaurs. Extant crocodilians, and folded into a series of tubes and for example, have as many as four addi­ chambers mounted above the rest of the tional diverticula of the nasal cavity prop­ skull (Fig 2-6). Hadrosaurids retain the er, although no single species has them antorbital sinus and its bony cavity, but it all (Witmer, 1995b, and references there­ is smaller and internalized as a result of in). These are best developed in alliga­ expansion of the nasal vestibule and its torines (see Fig 2-4) and pneumatize the recesses as well as expansion of the den­ maxillary, palatine, and prefrontal bones. tition and its supporting structures (Witmer, 1997), Other dinosaurs, such as anky­ Nasopharyngeal duct losaurids, may also have paranasal sinus­ Crocodilians also show pneumaticity es that derive from vestibular diverticula. associated with the nasopharyngeal duct. Certainly, ankylosaurids have a variety of In fact, six different pneumatic recesses paranasal sinuses, and at least one of derived from the nasopharygeal duct can them appears to communicate with the be recognized (Witmer, 1995b), involving nasal vestibule (Coombs and Maryanska, the pneumatization of the vomer, palatine, 1990; Witmer, 1997). and pterygoid bones. In many cases, large, inflated bullae are produced, such

30 Discussion

Discussion from the olfactory portion of the nasal cavity proper and involves the ethmoidal Paranasal air sinuses, although common system and its derivatives (e.g., frontal, in the more conspicuous terrestrial verte­ ethmoidal, and sphenoidal sinuses); eth­ brates such as birds and mammals, do moidal pneumaticity has a scattered not have a universal-or even a wide-dis­ distribution among eutherians but may be tribution among Vertebrata. In fact, the present in some nonmammalian synap­ only extant clades that definitely can be sids (Luo Z., personal communication, said to have true paranasal pneumaticity 1996). are eutherian mammals and archosaurs. Whereas mammalian paranasal pneu­ It seems almost certain that at least some maticity is restricted to diverticula arising synapsids more basal than Eutheria had from the nasal cavity proper, archosaurs some form of paranasal pneumaticity, but show examples of diverticula arising from the published distribution is difficult to all portions of the nasal cavity. The antor­ interpret and homologies are very bital sinus is the major (and usually only) unclear; a review of the evolution of paranasal air sinus arising from the nasal paranasal pneumaticity in Synapsida is in cavity proper, although this sinus often order. has a number of subsidiary diverticula Reconstruction of the phylogenetic that may pneumatize surrounding bones. history of paranasal pneumaticity in Crocodilians are remarkably pneumatic archosaurs has been a major focus of my animals and exhibit additional diverticula research for some time, and patterns are of the nasal cavity proper and an even beginning to emerge (Witmer, 1995b, greater diversity of diverticula of the 1997). Virtually all known taxa of nasopharyngeal duct. Sinuses arising as Archosauria, extinct as well as extant, diverticula of the nasal vestibule are very manifest at least some evidence of the rare, but the cavities within the narial crest presence of the antorbital air sinus. of lambeosaurine dinosaurs may be clas­ Moreover, it seems likely that this sified as such. paranasal air sac characterizes a slightly The status of paranasal sinuses in taxa more inclusive group, a group known as other than mammals and archosaurs is a Archosauriformes that includes a variety little more problematic. Certainly, there of basal forms ("thecodonts" such as are no known unequivocal examples of Euparkeria) as well as Archosauria itself. bony paranasal pneumatic recesses in All mammalian paranasal pneumaticity fishes, amphibians, turtles, or lepidosaurs. derives from diverticula from the nasal Nevertheless, thin-walled, epithelial, extra­ cavity proper. These diverticula can be capsular diverticula of the nasal cavity divided roughly into two systems. The first have been described in a number of system derives from the respiratory por­ these taxa (see Fig 2-2), and in mammals tion of the nasal cavity proper and involves and archosaurs it is just such diverticula the maxillary sinus; this sinus may well be that are involved in pneumatization of common to all Eutheria, with the excep­ osseous tissue. tion of a few taxa that apparently have lost What, then, is special about mammals it. The second system derives essentially and archosaurs that allows their diverticu-

31 The Phylogenetic History of Paranasal Air Sinuses la to playa role in pneumatization? The including those that have extracapsular answer is not at all clear. It might seem at diverticula but lack bony recesses, are first that physiologic considerations, such essential for elucidating gross aspects of as endothermy, may explain why mam­ the process of pneumatization. mals and birds are so pneumatic, but Perhaps most interesting would be crocodilians are ectothermic and are, by mechanistic studies of how the process of some measures, the most pneumatic of pneumatization is controlled. What factors all vertebrates. The fact is that very little is effect initiation of evagination of an epithe­ known about the processes of pneumati­ lial diverticulum? What controls its subse­ zation at the cellular and tissue levels, and quent growth? What determines whether even less is known about how pneumati­ or not such a diverticulum actually pneu­ zation is controlled. In any event, one con­ matizes bone? What is the relationship of clusion emerging from this situation is that the pneumatic epithelium to the front of the distinction between epithelial divertic­ osteoclasts that are actually involved in ulum and bony recess discussed at the the business of pneumatization? Once outset of the chapter is more than a these and other similar questions have semantic issue but rather is integral to satisfactory answers, then new ideas on understanding the nature of pneumaticity. the function of pneumatic sinuses (e.g., Witmer, 1997) will have appropriate tests, and the interpretation of the phylogenetic history of paranasal sinuses will be on a better footing. Conclusion

The phylogenetic history of paranasal air sinuses suggests that pneumaticity is a fairly poorly constrained system (Witmer, Acknowledgments 1997). Sinuses appear repeatedly and independently, and they are lost just as For useful discussion I thank Luo Zhexi frequently. In other words, the anatomic and John Ruben. system is characterized by a high level of Funding was provided by the Ohio homoplasy. The directions of future University College of Osteopathic research are clear. New paleontologic and Medicine, and the National Science comparative anatomic studies of particu­ Foundation, grants BSR-9112070 and lar clades of vertebrates-such as extant IBN-9601174. mammals, nonmammalian cynodonts, and nonamniote anthracosaurs-are needed, not only to establish the pres­ ence or absence of paranasal air sinuses but also to acquire sufficient anatomic res­ olution to shed light on the issue of homology. Furthermore, microanatomic and ontogenetic studies of extant taxa,

32 References

References Negus, V. E. The Comparative Anatomy and Physiology of the Nose and Paranasal Sinuses. London: E. &S. Livingstone. 1958. Coombs. W. P., and Maryanska, T. Ankylosauria. Nemours, P. R. Studies on the accessory nasal pp. 456-483. In Weishampel, D. B., Dodson, P., and sinuses: The comparative morphology of the nasal Osm6lska, H. (eds.) The Dinosauria. Berkeley, CA: cavities of Amphibia. Ann. Otol. Rhinol. Laryngol. University of California Press, 1990. 39:542-562,1930a. Dieulafe, L. Les fosses nasales des vertebres (mor­ Nemours, P. R. Studies on the accessory nasal phologie et embryologie). J. Anat. Physiol. 41: sinuses: The comparative morphology of the nasal 102-112,1905a. cavities of reptiles and birds. Ann. Otol. Rhinol. Dieulafe, L. Les fosses nasales des vertebres (mor­ Laryngol. 39: 1086-1108, 1930b. phologie et embryologie). J. Anat. Physiol. 41: Novacek, M. J. Patterns of diversity in the mam­ 300-318, 1905b. malian skull. pp. 438-545. In Hanken, J., and Hall, Dieulafe, L. Les fosses nasales des vertebres (mor­ B. K. (eds.) The Skull, vol. 2. Patterns of Structural phologie et embryologie). J. Anat. Physiol. 41: and Systematic Diversity. Chicago: University of 478-560, 1905c. Chicago Press, 1993.

Dieulafe, L. Les fosses nasales des vertebres (mor­ Parsons, 1. S. The nose and Jacobson's organ. pp. phologie et embryologie). J. Anat. Physiol. 41: 99-191. In Gans, C., and Parsons, T. S. (eds.) 658-680, 1905d. Biology of the Reptilia, vol. 2. New York: Academic Press, 1970. Gaffney, E. S. The cranial morphology of the extinct horned turtle, Meiolania platyceps, from the Paulli, S. Ober die Pneumaticitat des Schadels bei Pleistocene of Lord Howe Island, Australia. Bull. den Saugerthieren. Eine morphologische Studie. I. Am. Mus. Nat. Hist. 175: 361-480, 1983. Ober den Bau des Siebbeins. Ober die Mor­ phologie des Siebbeins und die der Pneumaticitat Gauthier, J. A., Kluge, A. G., and Rowe. 1. Amniote bei den Monotremen und den Marsupialiern. phylogeny and the importance of fossils. Clad. 4: Gegenb. Morphol. Jahrb. 28: 147-178, 1900a. 105-209, 1988. Paulli, S. Ober die Pneumaticitat des Schadels bei Jurgens, J. D. The morphology of the nasal region den Saugerthieren. Eine morphologische Studie. II. of Amphibia and its bearing on the phylogeny of Ober die Morphologie des Siebbeins und die der the group. Ann. Univ. Stellenb. 46: 1-146, 1971. Pneumaticitat bei den Ungulaten und Probosciden. Kalin, J. Crocodilia. pp. 695-784. In Piveteau, J. Gegenb. Morphol. Jahrb. 28: 179-251, 1900b. (ed.) Traite de Paleontologie, vol. V. Amphibiens, Paulli, S. Ober die Pneumaticitat des Schadels bei Reptiles, Oiseaux. Paris: Masson et Cie, 1955. den Saugerthieren. Eine morphologische Studie. Malan, M. E. Contributions to the comparative III. Ober die Morphologie des Siebbeins und die anatomy of the nasal capsule and the organ of der Pneumaticitat bei den Insectivoren, Hyra­ Jacobson. Ann. Univ. Stellenb. 24A: 69-137, 1946. coideen, Chiropteren, Carnivoren, Pinnipedien, Edentaten, Rodentieren, Prosimieren und Prima­ Matthes, E. Geruchsorgan. pp. 879-948. In Bolk, ten, nebst einer zusammenfassenden Obersicht L., Geppert, E., Kallius, E., and Lubosch, W. (eds.) uber die Morphologie des Siebbeins und die der Handbuch der Vergleichenden Anatomie der Pneumaticitat des Schadels bei den Saugerthieren. Wirbeltiere. Berlin: Urban &Schwarzenberg, 1934. Gegenb. Morphol. Jahrb. 28: 483-564, 1900c. Mihalkovics, V. Nasenhehle und Jacobsonsches Plate, L. Allgemeine Zoologie und Abstam­ Organ. Eine morphologische Studie. Anat. Hefte mungslehre. Jena, Germany: Gustav Fischer, 1924. 11: 1-107, 1898. Weishampel, D. B. The nasal cavity of lam­ Moore, W. J. The Mammalian Skull. Cambridge, beosaurine hadrosaurids (Reptilia: Ornithischia): England: Cambridge University Press, 1981. Comparative anatomy and homologies. J. Paleontol. 55: 1046-1057, 1981.

33 The Phylogenetic History of Paranasal Air Sinuses

Weishampel, D. B., and Horner, J. R. Ha­ Witmer, L. M. Homology of facial structures in drosauridae. pp. 534-561. In Weishampel, D. B., extant archosaurs (birds and crocodilians), with Dodson, P., and Osm6lska, H. (eds.) The special reference to paranasal pneumaticity and Dinosauria. Berkeley, CA: University of California nasal conchae. J. Morphol. 225: 269-327, 1995b. Press, 1990. Witmer, L. M. The evolution of the antorbital cavity Witmer, L. M. The craniofacial air sac system of of archosaurs: A study in soft-tissue reconstruction Mesozoic birds (Aves). ZooI. J. Linn. Soc. 100: in the fossil record with an analysis of the function 327-378, 1990. of pneumaticity. Mem. Soc. Vert. Paleontol. 17 (suppl.): 1-73, 1997. Witmer, L. M. The extant phylogenetic bracket and the importance of reconstructing soft tissues in fossils. pp. 19-33. In Thomason, J. J. (ed.) Functional Morphology in Vertebrate Paleontology. New York: Cambridge University Press, 1995a.

34