An d r z e j El z a n o w s k i 1,3 a n d Th o m a s A. St i d h a m 2

The Auk 128(1):138 145, 2011 The American Ornithologists Union, 2011. Printed in USA. A Galloanserine Quadrate from the Late Cretaceous Lance Formation of Wyoming An d r z e j El z a n o w s k i 1,3 a n d Th o m a s A. St i d h a m 2 1 Museum and Institute of Zoology, Polish Academy of Sciences, 64 Wilcza Street, 00-679 Warszawa, Poland; and 2 Department of Biology, Texas A&M University, 3258 TAMU, College Station, Texas 77843-3258, USA Abstract. Although the monophyly and Cretaceous origins of galloanserines have been established beyond a reasonable doubt, no stem galloanserine has ever been identified and very few fossils help to date the early phylogeny of the clade. We describe a very late Cretaceous quadrate that was previously assigned to Cimolopteryx, a charadriiform genus, and identify it instead as galloanserine and probably stem anseriform. In addition to the presence of a distinctive galloanserine character complex, the quadrate shows a predominance of plesiomorphic characters that are widespread among the Neornithes and shared by Presbyornis and primitive galliforms, in particular the megapodiids. Only one character, the unique configuration of the mandibular condyle, is known only in the anseriforms and thus indicates that the quadrate most likely belongs to a stem anseriform. The bone comes from a bird that compares in size to the smallest living galliforms, which suggests that early anseriforms, and possibly all galloanserines, were small. Their size may be responsible for the paucity of their Cretaceous record and, thus, its incongruence with the molecular dating of the anseriform galliform divergence. Received 30 April 2010, accepted 10 September 2010. Key words: Anseriformes, Cretaceous, Galliformes, Galloanseres, osteology, quadratum, skull. Un Cuadrado de un Galloanserino del Cretácico Tardío de la Formación Lance de Wyoming Resumen. Aunque la monofilia y el origen cretácico de los galloanserinos han sidos establecidos más allá de la duda razonable, ningún representante del tallo (i.e. stem group en inglés) del grupo de los galloanserinos ha sido identificado y muy pocos fósiles ayudan a datar la filogenia temprana del clado. Describimos un cuadrado del Crétacico muy tardío que anteriormente había sido asignado al género charadriforme Cimolopteryx y lo identificamos, en cambio, como perteneciente a un galloanserino y probablemente a un representante del tallo de los anseriformes. Además de la presencia de un conjunto de rasgos característico de los galloanserinos, el cuadrado muestra una predominancia de rasgos plesiomórficos que están ampliamente difundidos entre los Neornithes y son compartidos por Presbyornis y galliformes primitivos, particularmente los megapódidos. Solamente un carácter, la configuración única del cóndilo mandibular, se conoce sólo en los anseriformes, lo cual indica que lo más probable es que el cuadrado pertenezca a un representante del tallo de los anseriformes. El hueso proviene de una ave de tamaño comparable al de los galliformes vivientes más pequeños, lo que sugiere que los anseriformes tempranos (y posiblemente todos los galloanserinos) eran pequeños. Su tamaño podría ser la explicación de la escasez de fósiles cretácicos y, por tanto, de la incongruencia entre el registro fósil y la fecha de divergencia entre anseriformes y galliformes establecida mediante datos moleculares. The monophyly of galloanserines (Galloanseres, Galloanserae, Galloanserimorphae) is well supported by both morphological studies (Andors 1992, Weber 1993, Dzerzhinsky 1995, Zusi and Livezey 2000, Mayr and Clarke 2003, Livezey and Zusi 2007) and molecular phylogenetics (Ericson et al. 2006, Hackett et al. 2008). Molecular divergence estimates (van Tuinen 2009) place the Cretaceous split between the Anseriformes and Galliformes lineages 12 to 14 million years after the split between Galloanseres and Neoaves, leaving that period for the stem galloanserines to evolve in the absence of their modern descendants (see also Ksepka 2009). However, no stem galloanserine fossils have ever been identified, although some of them may have been assigned mistakenly to either galliforms or anseriforms. Basal taxa are notoriously difficult to identify with fragmentary evidence (van Tuinen et al. 2006, Mayr 2009). This is because fragments may show a predominance of either derived character states that today identify one of the crown groups, or plesiomorphic states that do not reveal any closer relationship. The presence of the anseriform lineage in the Late Cretaceous seems to be well established, although the phylogenetic status of Cretaceous anseriform fossils within the clade remains uncertain. The unquestionably anseriform Presbyornithidae, best known from the Eocene, have been identified in the Late Cretaceous deposits 3 E-mail: elzanowski@miiz.waw.pl The Auk, Vol. 128, Number 1, pages 138 145. ISSN 0004-8038, electronic ISSN 1938-4254. 2011 by The American Ornithologists Union. All rights reserved. Please direct all requests for permission to photocopy or reproduce article content through the University of California Press s Rights and Permissions website, http://www.ucpressjournals. com/reprintinfo.asp. DOI: 10.1525/auk.2011.10113 138

January 2011 Cretaceous Galloanserine Quadrate 139 (Hope 2002, Kurochkin et al. 2002), and their appendicular skeleton, in particular the wing skeleton and pectoral girdle, exhibits more than 10 apparently symplesiomorphic similarities (Ericson 2000:16 17) to early neoavian taxa, in particular the Cretaceous form family Graculavidae (Olson and Parris 1987, Boles 1999), which has traditionally been included among the charadriiforms. On the basis of such similarities, as well as the early fossil record of the Presbyornithidae, they have been interpreted as basal anseriforms with charadriiform affinities (Olson and Feduccia 1980). A recent study of the morphology of their quadrate supports their basal position among the anseriforms, but not their charadriiform similarities (Elzanowski and Stidham 2010). The plesiomorphic similarities of presbyornithids in both postcranial and quadrate morphology are in conflict with recent cladistic analyses (Ericson 1997, Livezey 1997, Dyke 2001, Clarke et al. 2005, Mayr 2008) that nest them within the crown anseriforms as a sister group to the Anatidae sensu stricto (to the exclusion of Anseranas and Anatalavis). However, the currently accepted phylogeny stems from the consensus between Ericson (1997) and Livezey (1997), because all three remaining studies (Dyke 2001, Clarke et al. 2005, Mayr 2008) used Livezey s (1997) matrix, in which the coding of some cranial characters supposedly shared by Presbyornis and Anatidae proved inaccurate or dubious (Elzanowski and Stidham 2010). The Late Cretaceous Antarctic fossil Vegavis has been championed as the best paleontological indication of the presence of crown anseriforms in the Cretaceous (Clarke et al. 2005). That bird is presbyornithid-like in the preserved postcranial morphology but differs from Presbyornis in skeletal proportions (Clarke et al. 2005). However, Vegavis exhibits the same states as Presbyornis in 10 out of 11 applicable characters of Livezey s (1997) matrix (characters 59, 70, 72, 78, 79, 82, 89, 90, 92, 94), which makes the placement of Vegavis almost totally dependent on and thus no more reliable than the placement of Presbyornis. Thus, Vegavis may confirm the presence of presbyornithid-like anseriforms in the Late Cretaceous but bring little new knowledge on the timing of anseriform phylogeny. In fact, if the deeply recessed plantar opening of the distal vascular foramen is unique to the Anatidae (Livezey 1997: character 92), then a likely candidate for the oldest modern anseriform fossil is the latest Cretaceous or earliest Paleocene tarsometatarsus from the Hell Creek Formation (Elzanowski and Brett-Surman 1995: fig. 2e). Cretaceous origins of the crown-group galliforms have been suggested by the Gondwanan distribution of the Megapodiidae and Cracidae (Cracraft 2001) and have apparently been corroborated by the phylogeography of galliforms, on the basis of both molecular and morphobehavioral data (Crowe et al. 2006) as well as the molecular divergence estimates based on mitochondrial sequences (Pereira and Baker 2006). However, the latter two studies used misclassified fossils for calibrating nodes, so the resulting hypotheses are likely flawed (Mayr 2008, 2009: p. 22ff.). Most significantly, the Cretaceous record of the galliforms remains remarkably poor. The strongest piece of evidence is the distal fragment of the tarsometarsus of Austinornis lentus (Clarke 2004) from the Late Cretaceous Austin Chalk of Texas. Hope (2002) identified as galliform the coracoids and scapula of Palintropus. However, Longrich (2009) suggested that this genus had a close relationship to the early ornithurine Apsaravis, and Mayr (2009:19) concluded that Palintropus could at best be a sister taxon to all other galliforms. Also, galliform affinities have been suggested for a partial wing skeleton from the Campanian Two Medicine Formation of Montana (Varricchio 2002) and a coracoid from the much older, Turonian- Coniacian Portezuelo Formation of Patagonia (Agnolin et al. 2006), but none of those fragmentary remains can definitively establish the presence of galliforms in the Cretaceous. The oldest definitive representative of the galliform lineage is the early Eocene Gallinuloides (Ksepka 2009, Mayr 2009), and the first crowngroup galliforms are known only in the Oligocene (Mayr 2009). The late appearance of the galliforms may reflect a strong taphonomic bias against them, the relatively recent origins of modern galliform morphology, or their origin in Gondwana (which has a relatively poor Paleogene bird record). A previously described Cretaceous fossil may help to resolve some of the issues related to the early history of galloanserines. We redescribe a small quadrate from the latest Cretaceous Lance Formation of Wyoming (North America) and unequivocally assign that bone to the galloanserine clade. The Lance Formation in eastern Wyoming was deposited in a subtropical climate along a shoreline during the final retreat of the North American inland sea (Clemens 1963). Plant and vertebrate fossils are abundant in exposures of the ancient near-shore streambeds (Clemens 1963). The Lance Creek area localities (including Lull 2) are stratigraphically high within the formation (Clemens 1963). The Lance Formation does not stretch to the Cretaceous Paleogene Boundary in the Lance Creek area, and those outcrops contain magnetically reversed rock sequences that indicate their deposition during magnetochron 29R (Keating and Helsey 1983, Gradstein et al. 2004). Thus, the sediments and their fossils (including this quadrate) were deposited within the last 333,000 years of the end of the Cretaceous (Gradstein et al. 2004). Several avian fossils, Ceramornis major, Cimolopteryx rara, Lonchodytes estesi, L. pterygius, and Torotix clemensi, were described from the same quarry, and two other Cimolopteryx spp., Graculavus augustus, and Palintropus retusus from other nearby Lance Formation quarries (Brodkorb 1963, 1970; Hope 1999, 2002). All of those taxa and specimens were treated as primitive charadriiforms by Olson (1985), but the true taxonomic status of some of these specimens is not well established, and Palintropus may in fact represent another Lance Formation galloanserine, possibly a stem galliform (Hope 2002, Mayr 2009). The quadrate UCMP 53969 was described briefly by Brodkorb (1963), who referred it to Cimolopteryx rara, which is generally considered a plesiomorphic shorebird species (named from a coracoid). Unfortunately, the published description is incomplete and partly inaccurate, and the quadrate has since been damaged, rendering some of Brodkorb s observations unverifiable. Ours is the first complete description of this specimen based on its study in the present condition and using Brodkorb s previously unpublished photographs prior to the significant damage it experienced. The morphological states present in the fossil bring new data to the history of galloanserine birds and to avian diversity at the end of the Cretaceous. Material and Methods This specimen, University of California Museum of Paleontology at Berkeley (UCMP) 53969, was collected at UCMP Lance Formation locality V-5620 (Lull 2 quarry), near Buck Creek, a tributary

140 Elzanowski and Stidham Auk, Vol. 128 Fig. 1. Left quadrate UCMP 53969 (as recovered from P. Brodkorb s original photographic negative). (A) Caudal, (B) rostral, and (C) medial view. Abbreviations: bo = fossa basiorbitalis; cl = crista lateralis; cm = crista medialis; dc = depressio caudomedialis; dp = depressio praecondylaris; ds = depressio supracondylaris; fb = foramen pneumaticum basiorbitale; fm = foramen pneumaticum rostromediale; l = condylus mandibularis lateralis; lb = rostral labrum of the mandibular condyles; m = condylus mandibularis medialis; o = capitulum oticum; pc = condylus pterygoideus; pq = pars quadratojugalis of the lateral process; qs = fovea quadratojugalis; s = capitulum squamosum; t = tuberculum subcapitulare. Osteological terminology from Elzanowski and Stidham (2010). Scale bar = 3 mm. of Lance Creek, Niobrara County, Wyoming. UCMP 53969 is a left quadrate that was fairly complete at the time of Brodkorb s study except for the orbital process (Fig. 1). Subsequently, the entire medioventral part of the bone, including the pterygoid and medial mandibular condyles, has been damaged. Our description is based on the specimen after this damage, as well as on unpublished photographs (as recovered from the original photographic negative) that were taken by the late Pierce Bordkorb (University of Florida at Gainesville) around the year 1978. Osteological terminology for the quadrate follows Elzanowski and Stidham (2010). For morphological comparisons, we used the data set from 33 genera in all eight extant gallaoanserine families as in our study of Presbyornis (Elzanowski and Stidham 2010) and the survey of all neoavian nonpasserine families (Elzanowski et al. 2000). In addition, we measured height, capitular span, and condylar span (Table S1 in online Supplemental Materials; see Acknowledgments) from one or two species from each of the following 26 families: Columbidae, Podicipedidae, Phaethontidae, Burhinidae, Charadriidae, Laridae, Alcidae, Gruidae, Diomedeidae, Procellariidae, Pelecanidae, Scopidae, Ardeidae, Ciconiidae, Fregatidae, Phalacrocoracidae, Spheniscidae, Gaviidae, Cathartidae, Accipitridae, Falconidae, Cariamidae, Strigidae, Meropidae, Bucerotidae, and Sturnidae. We compared all three dimensions simultaneously in a ternary diagram generated in SIGMAPLOT, version 11.2. 14_Elzanowski_10-113.indd 140 D escription The quadrate UCMP 53969 of the newly identified Lance galloanserine shows smooth bone surfaces (including the articular facets), which indicates that the quadrate is from an adult or subadult individual. The quadrate is 6.0 mm tall and thus slightly (0.5 mm) shorter than the quadrate of the 162- to 170-g Gambel s Quail (Callipepla gambelii) and much (2.3 mm) shorter than the bone in the smallest living anatid, the 260- to 285-g African Pygmy-Goose (Nettapus auritus). On the assumption that the quadrate-versusbody-mass allometry does not vary substantially within both the galliforms and the anseriforms, this suggests that the body mass of the new Lance galloanserine is in the range of 150 210 g. Because the distal tarsometatarsus of Gambel s Quail (5.5 6.2 mm of width) is much narrower (by ~4 mm) than that of either the likely anseriform (UCMP 117599) from the Hell Creek Formation (Elzanowski and Brett-Surman 1995) or that of the putative galliform Austinornis lentus (Clarke 2004) from the much older Austin Chalk, the new Lance galloanserine was much smaller than either of these likely galloanserines. It also was much smaller than Vegavis and Presbyornis, both of which have much longer leg bones than the smallest anatids. The two capitula are widely separated (Figs. 1A, B and 2A) and span 2.55 mm. There is little difference between the capitula 2/10/11 12:44:06 PM

J anuary 2011 C retaceous G alloanserine Q uadr ate 141 Fig. 2. Left quadrates in dorsal (proximal) view of the capitula (both with the rostral side up): (A) UCMP 53969, (B) Alectura lathami, (C) Crax alector, (D) Anhima cornuta. In Presbyornis the otic capitulum is slightly larger than the squamosal capitulum. In the Phasianoidea, the otic capitulum is even smaller than in the Cracidae. Abbreviations are as in Figure 1, except ip = intecapitular vallecula. in terms of their proximal (dorsal) extension, the squamosal capitulum extending at most only slightly more than the otic one. The otic capitulum is a rounded trapezoid in outline with a maximum diameter of 0.9 mm, and its articular facet is larger (contra Brodkorb 1963) than the squamosal capitulum (which is elongate, 0.9 0.5 mm). The squamosal facet is biplanar, with a distinct rostral slope. The medial margin of the squamosal facet projects dorsally beyond the otic facet. The squamosal facet is more slanted (laterally) than the otic facet (medially). Both the lateral crest (which connects the squamosal capitulum to the base of the lateral process) and the medial crest (which begins ~1 mm below the otic capitulum and extends to the pterygoid condyle) are well developed (Fig. 1A, B, C). The lateral crest is in a laterocaudal position in accordance with the caudal slant of the lateral process, which is typical of all galloanserines. The tympanic crest is barely marked ventrally and seems confluent with the lateral crest dorsally (a configuration reminiscent of the Anhimidae). Just below the squamosal capitulum, there is a subcapitular tubercle that is narrow, elongate dorsoventrally, and slightly concave rostrally and convex caudally (Fig. 1A, B). Just below the otic capitulum and above the medial crest, there is a minute flat area with an indication of the slit-like, vertical, 0.7-mm-long foramen, which is possibly the rostromedial foramen (Figs. 1C and 3A). Another, even smaller, barely noticeable foramen is present rostral to the medial crest, at the mid-height of the bone. 14_Elzanowski_10-113.indd 141 Below the squamosal capitulum there are shallow depressions in both the rostral and the caudal surfaces of the bone. A major feature of the caudal aspect is an elongate caudomedial depression (now damaged) that deepens ventrally but does not contain a foramen. The tympanic crest that, by definition, defines the depression laterally is rounded and inconspicuous. The basiorbital fossa (Fig. 1C) contains an elongate basiorbital foramen: a large pneumatic foramen medial to the base of the orbital process (Brodkorb 1963:67). There is no indication of the orbitocotylar crest. An intact medial process was present before the specimen was damaged (Fig. 1). The prominent pterygoid condyle is separated from the medial condyle by a wide incisura intercondylaris. There is no indication that the pterygo-quadrate articulation extends to the base of the orbital process (Fig. 1B), but it cannot be ruled out. The pterygoid condyle was described by Brodkorb (1963) as flat, unswollen, two-pronged, which is incomprehensible when compared with Brodkorb s own photograph (Fig. 1), and thus that statement seems to be a mistake. As in all galloanserines, the entire lateral process slants caudally, whereby the quadratojugal articulation is situated much more caudally than the pterygoquadrate articulation (Fig. 1C). The caudal slant of the lateral process is emphasized by the large pars quadratojugalis that is distinct from the lateral condyle. The fovea quadratojugalis is moderately deep and has an internal diameter of 0.6 mm. The quadratojugal articulation surfaces are poorly 2/10/11 12:44:09 PM

142 Elzanowski and Stidham Auk, Vol. 128 Fig. 3. The quadrates in medial views (A) UCMP 53969, (B) Megapodius freycinet, and (C) Presbyornis (as reconstructed by Elzanowski and Stidham 2010). Abbreviations are as in Figure 1, except fc = foramen pneumaticum caudomediale; or = processus orbitalis; po = facies articularis pterygoidea of the orbital process. differentiated, the caudal incisure is absent, and the cotyla merges with the quadrate body in the area of the rostral incisure. The maximum ventral width of the pars mandibularis, between the ends of the pterygoid condyle and quadratojugal socket, is 3.6 mm. The entire intact mandibular articulation is 2.8 2.9 mm wide (condylar width). The rostral margin of the mandibular surface projects as a rostral labrum that ventrally bounds the supracondylar depression (Fig. 1B). Dorsally, the supracondylar depression extends up to the base of the orbital process. Lateral to the basiorbital fossa, enclosed between the labrum and the quadratojugal socket, is a distinct pit, which corresponds to the precondylar depression in other birds (Elzanowski et al. 2000). The mandibular surface has a concave rostral profile and a convex caudal profile (Fig. 4B). The ventral profile (as seen in rostral or caudal view) is concave in the middle (Fig. 1A, B), but this concavity does not really represent the intercondylar vallecula. The two condyles have their beveled tips overlapping in the middle, leaving a narrow and shallow intercondylar vallecula that is oblique with respect to the plane connecting the condyles. The tip of the medial condyle reaches behind the tip of the lateral condyle but does not form any caudal expansion corresponding to the caudal condyle. The lateral condyle is 0.9 mm wide (rostrocaudally), convex (rounded), and much larger than the now damaged medial condyle that was 0.6 mm wide. Judging from the photograph (Fig. 1A, C), the medial condyle may have borne a narrow articular surface facing ventromedially in addition to the main articular surface facing ventrolaterally. D iscussion Brodkorb (1963) claimed that quadrate UCMP 53969 had similarities to charadriiforms (especially Recurvirostra) and referred it, along with other skeletal elements from the Lance Formation 14_Elzanowski_10-113.indd 142 Fig. 4. Left quadrates in ventral (distal) view of the condyles (both with the rostral side up): (A) Alectura lathami, (B) UCMP 53969, and (C) Presbyornis. Abbreviations are as in Figure 1, except in = intecondylar vallecula. 2/10/11 12:44:11 PM

January 2011 Cretaceous Galloanserine Quadrate 143 Fig. 5. Ternary plot (with an explanatory diagram on the left) showing proportions of the quadrate in galloanserines and other euornithine birds. See text for the list of families and Table S1 in the Supplementary Material for the list of species, measurements, and numeric ratios. (i.e., a carpometacarpus fragment), to Cimolopteryx rara, which was originally based only on a coracoid (Marsh 1892). Family Cimolopterygidae is believed to include primitive charadriiforms (Olson 1985, Hope 2002), but Hope (2002:372) listed the quadrate UCMP 53969 among Neornithes incertae sedis and interpreted its morphology as plesiomorphic among neornithines rather than charadriiform. We agree that the quadrate does not show a single specific similarity to the Charadriiformes in general and to Recurvirostra in particular. In fact, it exhibits a combination of uniquely galloanserine (Elzanowski and Stidham 2010) and plesiomorphic neornithine (Elzanowski et al. 2000) characters. The quadrate is clearly euornithine (ornithurine), as revealed primarily by its prominent pterygoid condyle. In contrast to the bicondylar quadrates in Archaeopteryx and the enantiornithines (Gobipteryx), the galloanserine bicondylar condition, as found in UCMP 53969, is highly derived, the two condyles being so close to each other that their articular facets are contiguous and often continuous, whereby the entire mandibular articulation becomes relatively narrow (Fig. 5). Among the euornithines, a bicondylar quadrate (without any trace of a caudal condyle) also evolved in the pigeons (Columbidae), whose two simple, widely spaced condyles are more reminiscent of the primitive Archaeopteryx-like condition. The galloanserine character complex of the quadrate includes a mandibular articulation that is bicondylar, without any trace of the caudal condyle, and narrower than half the height of the quadrate (Fig. 5); the entire lateral process (the lateral condyle and the quadratojugal part) directed laterocaudally; the condyles aligned from rostromedial to caudolateral (Fig. 4), thus enabling rostrocaudal sliding movements in the mandibular joint (only in the majority of phasianids is the medial condyle set off caudally, conveying a comma shape to the entire mandibular articulation); and the presence of a subcapitular tubercle (which also occurs in other clades, but not in combination with the characters listed above). The galloanserines share general proportions of the quadrate with the odontognaths but differ (without any overlap) from all other neornithines (Fig. 5). Galloanserine and odontognathous quadrates are much slimmer than those of either neognaths or palaeognaths, primarily because of the relative height of the bone and the narrowness of the mandibular articulation (Fig. 5). Although the quadrates of Anatidae and Anseranas differ greatly from those of nonmegapodiid galliforms, the quadrates of megapodiids, anhimids, and Presbyornis are very close to each other, and thus likely to show plesiomorphic galloanserine proportions. UCMP 53969 is closest to the plesiomorphic cluster in the relative height and condylar span, but differs in having a larger capitular span (dorsal width) that is almost exactly the average for non-galloanserine neornithines (Fig. 5). As a likely primitive state, this is consistent with the tendency toward the integration and narrowing of the quadrate head in the galliforms and, to a lesser extent, the anseriforms. The quadrate UCMP 53969 differs in at least four characters from most crown-group galloanserines and retains a plesiomorphic neornithine condition, also observed in Presbyornis (Elzanowski and Stidham 2010): (1) two pneumatic foramina, basiorbital and rostromedial or caudomedial (Figs. 1C and 3); (2) a prominent pterygoid condyle (Fig. 1) that is widely separated from either the medial condyle (by the intercondylar incisure) or the orbital process; (3) the capitula widely separated from each other, in fact wider than in any other galloanserine (Figs. 2 and 5); and (4) the quadratojugal socket completely closed (Fig. 1B) rather than open caudoventrally as in the extant galloanserines (except for the odontophorids). Among extant galloanserines, the presence of two pneumatic foramina, a large basiorbital foramen and a small rostromedial foramen in a dorsal (subcapitular) position, is characteristic of the galliforms (in some of which the rostromedial foramen disappears altogether). However, the presence of two pneumatic foramina, basiorbital and rostromedial or caudomedial (both receiving probably the same pneumatic diverticulum), is widespread among the neovians (Elzanowski and Stidham 2010), so this condition is likely primitive. Among the galliforms, the quadrate UCMP 53969 is most similar to that of the Megapodiidae, especially Megapodius (Fig. 3), primarily in proportions (Fig. 5) and the configuration of the head with little or no difference in the dorsal extension of the capitula

144 El z a n o w s k i a n d St i d h a m Au k, Vo l. 128 and the squamosal capitulum no larger than the otic one (Fig. 2). All these similarities are likely to be symplesiomorphic for the galloanserines, because the megapodiids are the sister group to all other crown galliforms (Mayr and Weidig 2004, Crowe et al. 2006, Pereira and Baker 2006) and their postcranial skeletons closely approach in morphology the Paleogene stem-galliform family Quercymegapodiidae (Mourer-Chauviré 1992). The specific similarities of the Lance fossil to Megapodius are also likely symplesiomorphic, because the genera Megapodius and Eulipoa probably constitute the sister group to all other megapodiids (Birks and Edwards 2002) and seem to be the least specialized, at least in terms of reproductive behavior. Within the galloanserines, the quadrate UCMP 53969 shares at least two characters with the anseriforms: a distinct caudomedial depression and a specific configuration of the mandibular condyles. The caudomedial depression, between the medial and tympanic crests, is widespread among the neoavians and thus may well be a plesiomorphy retained by the anseriforms and reduced in the galliforms. By contrast, the subparallel orientation of the mandibular condyles that interdigitate with their beveled tips in such a way that the medial condyle extends caudally to the lateral condyle (Fig. 2) is a unique anseriform character (Elzanowski and Stidham 2010), indicating anseriform affinities of the taxon that yielded the quadrate UCMP 53969. In the galliforms, the condyles broadly slope into the vallecula without any well-defined margins (Fig. 2; see also Weber 1996: fig. 10c). Among the anseriforms, the quadrate UCMP 53969 is most similar to that of Presbyornis in the presence of a large basiorbital foramen, the absence of any dorsal expansions of the pars quadratojugalis, a wide intercapitular incisure (Fig. 5), the otic capitulum larger than the squamosal one (Fig. 2) (although it is much larger in UCMP 53969, but only slightly larger in Presbyornis), and a wide intercondylar incisure (that also remains relatively wide in the anhimids). All these similarities are anseriform symplesiomorphies and thus appear to support the recognition of the Presbyornithidae as stem anseriforms (Elzanowski and Stidham 2010). We conclude that the quadrate UCMP 53969 belongs certainly to a Cretaceous galloanserine and most probably to a stemgroup anseriform (near the split with the galliforms). Inasmuch as the quadrate is representative of the entire head morphology of the stem anseriforms, it must have been a mosaic of galliform and anseriforms characters with a prevalence of the former. The megapodiid similarities in both UCMP 53969 and Presbyornis (Elzanowski and Stidham 2010) suggest that the Megapodius-like morphology of the head may have been characteristic of the stem galloanserines. The crown-group galloanserines are, on average, much larger than UCMP 53969, which suggests that the evolution of gallonserines may have followed Cope s rule of phyletic size increase, which has been confirmed to prevail in both extant (Maurer 1998) and Mesozoic (Hone et al. 2008) avian lineages, and has been explained by showing that both natural and sexual selection favor larger body size (Kingsolver and Pfennig 2004). If stem galloanserines were indeed the size of the new Lance galloanserine (as represented by UCMP 53969), then their origin may have been obscured by the taphonomic bias against small birds under average depositional conditions, which may explain the discrepancy between the predictions from molecular phylogenies and the fossil record. Ac k n o w l e d g m e n t s Supplementary material (Table S1) is available at dx.doi.org/10.1525/ auk.2011.10113. We thank S. Hope (California Academy of Sciences) for Brodkorb s original photographic negative, from which Figure 1 was prepared; A. Manegold (Forschungsinstitut und Museum Senckenberg) for comparative osteological data; E. Weber (Universität Tübingen) and an anonymous reviewer for extensive comments; P. Holroyd for access to material and data in the University of California Museum of Paleontology; and Z. Bocheński for continuous access to the extant skeleton collection of the Institute of Animal Systematics and Evolution, Polish Academy of Sciences. Literature Cited Agnolin, F. L., F. E. Novas, and G. Lio. 2006. Neornithine bird coracoid from the Upper Cretaceous of Patagonia. Ameghiniana 43:245 248. Andors, A. V. 1992. Reappraisal of the Eocene groundbird Diatryma (Aves: Anserimorphae). Natural History Museum of Los Angeles County Science Series 36:109 125. Birks, S. M., and S. V. Edwards. 2002. A phylogeny of the megapodes (Aves: Megapodiidae) based on nuclear and mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 23:408 421. Boles, W. 1999. Early Eocene shorebirds (Aves: Charadriiformes) from the Tingamarra Laocal Fauna, Murgon, Queensland, Australia. Records of the Western Australian Museum Supplement 57:229 238. Brodkorb, P. 1963. Birds from the Upper Cretaceous of Wyoming. Pages 55 70 in Proceedings XIII International Ornithological Congress (C. G. Sibley, Ed.). American Ornithologists Union, Washington, D.C. Brodkorb, P. 1970. The generic position of a Cretaceous bird. Quarterly Journal of the Florida Academy of Sciences 32:239 240. Clarke, J. A. 2004. Morphology, phylogenetic taxonomy, and systematics of Ichthyornis and Apatornis (Avialae: Ornithurae). Bulletin of the American Museum of Natural History, no. 286. Clarke, J. A., C. P. Tambussi, J. I. Noriega, G. M. Erickson, and R. A. Ketcham. 2004. Definitive fossil evidence for the extant avian radiation in the Cretaceous. Nature 433:305 308. Clemens, W. A. 1963. Fossil mammals of the type Lance Formation, Wyoming, Part I. Introduction and Multituberculata. University of California Publications in Geological Sciences, no. 48. Cracraft, J. 2001. Avian evolution, Gondwana biogeography and the Cretaceous Tertiary mass extinction event. Proceedings of the Royal Society of London, Series B 268:459 469. Crowe, T. M., R. C. K. Bowie, P. Bloomer, T. G. Mandiwana, T. A. J. Hedderson, E. Randi, S. L. Pereira, and J. Wakeling. 2006. Phylogenetics, biogeography and classification of and character evolution in gamebirds (Aves: Galliformes): Effects of character exclusion, data partitioning and missing data. Cladistics 22:495 532. Dyke, G. J. 2001. The fossil waterfowl (Aves: Anseriformes) from the Eocene of England. American Museum Novitates 3354:1 15. Dzerzhinsky, F. Y. 1995. Evidence for common ancestry of the Galliformes and Anseriformes. Courier Forschungsinstitut Senckenberg 181:325 336.

January 2011 Cretaceous Galloanserine Quadrate 145 Elzanowski, A., and M. K. Brett-Surman. 1995. Avian premaxilla and tarsometatarsus from the Uppermost Cretaceous of Montana. Auk 112:762 767. Elzanowski, A., G. S. Paul, and T. A. Stidham. 2000. An avian quadrate from the Late Cretaceous Lance Formation of Wyoming. Journal of Vertebrate Paleontology 20:712 719. Elzanowski, A., and T. A. Stidham. 2010. Morphology of the quadrate in the Eocene anseriform Presbyornis and extant galloanserine birds. Journal of Morphology 271:305 323. Ericson, P. G. P. 1997. Systematic relationships of the Palaeogene family Presbyornithidae (Aves: Anseriformes). Zoological Journal of the Linnean Society 121:429 483. Ericson, P. G. P. 2000. Systematic revision, skeletal anatomy, and paleoecology of the New World early Tertiary Presbyornithidae (Aves: Anseriformes). PaleoBios 20:1 23. Ericson, P. G. P., C. L. Andersson, T. Britton, A. Elzanowski, U. S. Johansson, M. Källersjö, J. I. Ohlson, T. J. Parsons, D. Zuccon, and G. Mayr. 2006. Diversification of Neoaves: Integration of molecular sequence data and fossils. Biology Letters 2:543 547. Gradstein, F., J. Ogg, and A. Smith. 2004. A Geologic Time Scale 2004. Cambridge University Press, Cambridge, United Kingdom. Hackett, S. J., R. T. Kimball, S. Reddy, R. C. K. Bowie, E. L. Braun, M. J. Braun, J. L. Chojnowski, W. A. Cox, K. Han, J. Harshman, and others. 2008. A phylogenomic study of birds reveals their evolutionary history. Science 320:1763 1768. Hone, D. W. E., G. J. Dyke, M. Haden, and M. J. Benton. 2008. Body size evolution in Mesozoic birds. Journal of Evolutionary Biology 21:618 624. Hope, S. 1999. A new species of Graculavus from the Cretaceous of Wyoming (Aves: Neornithes). Smithsonian Contributions to Paleobiology 89:261 266. Hope, S. 2002. The Mesozoic radiation of Neornithes. Pages 339 388 in Mesozoic Birds: Above the Heads of Dinosaurs (L. M. Chiappe and L. W. Witmer, Eds.). University of California Press, Berkeley. Keating, B. H., and C. E. Helsey. 1983. The magnetostratigraphy of the Cretaceous Tertiary Boundary in the continental Lance Formation and five marine sequences. Eos 64(18):219. Kingsolver, J. G., and D. W. Pfennig. 2004. Individual-level selection as a cause of Cope s Rule of phyletic size increase. Evolution 58:1608 1612. Ksepka, D. 2009. Broken gears in the avian molecular clock: New phylogenetic analyses support stem galliform status for Gallinuloides wyomingensis and rallid affinities for Amitabha urbsinterdictensis. Cladistics 25:173 197. Kurochkin, E. N., G. J. Dyke, and A. A. Karhu. 2002. A new presbyornithid bird (Aves, Anseriformes) from the Late Cretaceous of southern Mongolia. American Museum Novitates 3386:1 11. Livezey, B. C. 1997. A phylogenetic analysis of basal Anseriformes, the fossil Presbyornis, and the interordinal relationships of waterfowl. Zoological Journal of the Linnean Society 121:361 428. Livezey, B. C., and R. L. Zusi. 2007. Higher-order phylogeny of modern birds (Theropoda, Aves: Neornithes) based on comparative anatomy. II. Analysis and conclusion. Zoological Journal of the Linnean Society 149:1 95. Longrich, N. 2009. An ornithurine-dominated avifauna from the Belly River Group (Campanian, Upper Cretaceous) of Alberta, Canada. Cretaceous Research 30:161 177. Marsh, O. C. 1892. Notes on Mesozoic vertebrate fossils. American Journal of Science series 3, no. 44:171 175. Maurer, B. A. 1998. The evolution of body size in birds. I. Evidence for non-random diversification. Evolutionary Ecology 12:925 934. Mayr, G. 2008. Phylogenetic affinities and morphology of the late Eocene anseriform bird Romainvillia stehlini Lebedinsky, 1927. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen, 248:365 380. Mayr, G. 2009. Paleogene Fossil Birds. Springer, Berlin. Mayr, G., and J. Clarke. 2003. The deep divergences of neornithine birds: A phylogenetic analysis of morphological characters. Cladistics 19:527 553. Mayr, G., and I. Weidig. 2004. The early Eocene bird Gallinuloides wyomingensis A stem group representative of Galliformes. Acta Palaeontologica Polonica 49:211 217. Mourer-Chauviré, C. 1992. The Galliformes (Aves) from the Phosphorites du Quercy (France): Systematics and biostratigraphy. Natural History Museum of Los Angeles County Science Series 36:67 96. Olson, S. L. 1985. The fossil record of birds. Pages 79 238 in Avian Biology, vol. 8 (D. S. Farner, J. R. King, and K. C. Parkes, Eds.). Academic Press, New York. Olson, S. L., and A. Feduccia. 1980. Presbyornis and the origin of the Anseriformes (Aves: Charadriomorphae). Smithsonian Contributions to Zoology 323:1 24. Olson, S. L., and D. C. Parris. 1987. The Cretaceous birds of New Jersey. Smithsonian Contributions to Paleobiology 63:1 22. Pereira, S. L., and A. J. Baker. 2006. A molecular timescale for galliform birds accounting for uncertainty in time estimates and heterogeneity of rates of DNA substitutions across lineages and sites. Molecular Phylogenetics and Evolution 38:499 509. van Tuinen, M. 2009. Birds (Aves). Pages 409 411 in The Timetree of Life (S. Blair Hedges and S. Kumar. Eds.). Oxford University Press. van Tuinen, M., T. A. Stidham, and E. A. Hadly. 2006. Tempo and mode of modern bi rd evolution observed with large-scale taxonomic sampling. Historical Biology 18:205 221. Varricchio, D. J. 2002. A new bird from the Upper Cretaceous Two Medicine Formation of Montana. Canadian Journal of Earth Sciences 39:19 26. Weber, E. 1993. Zur Evolution basicranialer Gelenke bei Vögeln, insbesondere bei Hühner- und Entenvögeln (Galloanseres). Zeitschrift für zoologische Systematik und Evolutionsforschung 31:300 317. Weber, E. 1996. Das Skelet-Muskel-System des Kieferapparates von Aepypodius arfakianus (Salvadori, 1877) (Aves, Megapodiidae). Courier Forschungsinstitut Senckenberg 189:1 132. Zusi, R. L., and B. C. Livezey. 2000. Homology and phylogenetic implications of some enigmatic cranial features in galliform and anseriform birds. Annals of Carnegie Museum 69:157 193. Associate Editor: G. Mayr