GouldQuitmyer 201 TITANIS WALLERI: BONES OF CONTENTION Gina C. Gould1 and Irvy R. Quitmyer2 Titanis walleri, one of the largest and possibly the last surviving member of the otherwise South American Phorusrhacidae is re- considered in light of all available data. The only verified phorusrhacid recovered in North America, Titanis was believed to exhibit a forward-extending arm with a flexible claw instead of a traditional bird wing like the other members of this extinct group. Our review of the already described and undescribed Titanis material housed at the Florida Museum of Natural History suggest that Titanis: (1) was like other phorusrhacids in sporting small, ineffectual ratite-like wings; (2) was among the tallest of the known phorusrhacids; and (3) is the last known member of its lineage. Hypotheses of its range extending into the Pleistocene of Texas are challenged, and herein Titanis is presumed to have suffered the same fate of many other Pliocene migrants of the Great American Interchange: extinction prior to the Pleistocene. Key Words: Phorusrhacidae; Great American Biotic Interchange; Florida; Pliocene; Titanis 1Gina C. Gould, Syncreta Associates, Inc., 6290 Grandview Court, Keystone Heights, FL 32656; 2Irvy R. Quitmyer, Environmental Archaeology, Florida Mu- seum of Natural History, Florida Museum of Natural History, PO Box 117800, University of Florida, Gainesville, Florida 32611- 7800; Bull. Fla. Mus. Nat. Hist. (2005) 45(4): 201-229 INTRODUCTION Titanis walleri (Brodkorb 1963), more commonly known as the North American ‘Terror Bird’, is one of the larg- est known phorusrhacids, an extinct group of flightless carnivorous birds from the Tertiary of South America, and most likely, the last known member of its lineage (Brodkorb 1967; Tonni 1980; Marshall 1994; Alvarenga & Höfling 2003). Titanis was first proposed as a par- ticipant in the Great American Interchange by David Webb and his colleagues (Marshall et al. 1982; Webb 1985), because it appears in Florida just after the forma- tion of the Panamanian land bridge (Stelhi & Webb 1985). In 1961, Titanis was discovered from a Blancan site in the Santa Fe River along the Gilchrist/Columbia County line in Florida (Brodkorb 1963; Fig. 1) circa David Webb’s arrival at the Florida Museum of Natural His- tory (FLMNH) as the new curator of vertebrate pale- ontology. At the time, the newly recovered material consisted of a distal end of a tarsometatarsus (the holo- type), a phalanx from digit III, and the proximal end of a fibula (never described). Based on the gigantic size of the fossils and the presence of a bifurcated distal fora- men on the tarsometatarsus, these specimens were as- signed to the Family Phorusrhacidae (Brodkorb 1963) and named after both a Titan Goddess from Greek my- thology and Benjamin Waller, the discoverer of the fos- sils (Zimmer 1997). Since then, isolated Titanis mate- rial has been recovered from three other localities in Florida (Table 1; Fig. 1; Brodkorb 1963; Carr 1981; Chandler 1994; Hulbert pers com.) and one in Texas (Baskin 1995). Despite considerable effort on the part of the FLMNH and enthusiasts to uncover more com- plete Titanis material, few new remains have been found in the now 30+year search (R. Hulbert pers comm., Table 1). Most of the material is fragmentary, conse- quently much of it remained undescribed. Regardless of the scant and incomplete nature of the Titanis material, the bird is well known, if not infa- mous among paleontologists and paleoenthusiasts. Titanis has been described as a giant flightless ‘terror’ bird between two and three meters tall (Brodkorb 1963; Marshall 1994; Feduccia 1999), with a “strong, robust wing, and an extended manus (as in penguins) equipped with a large claw which was used to subdue struggling prey” (Chandler 1994:176). Although published descrip- tions of this taxon are few (Brodkorb 1963; Chandler 1994; Baskin 1995; Emslie 1998; Alvarenga & Höfling 2003; and one dissertation, Carr 1981; see Table 1), Titanis has made its way into the popular literature. In one of the more prominent articles, Titanis was described to have had a “three-foot-long wing” which was more 202 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb like an “arm with a giant movable claw and two smaller fixed claws” (Zimmer 1997, Discover Magazine). The shear size of Titanis and the fantastic nature of its wing pose intriguing questions about the biology and ecology of this extinct bird. An accurate reconstruction of Titanis became even more acutely important to researchers and artists at the FLMNH when they wanted to display a full-scale model of it for their new exhibit hall, Florida Fossils: Evolution of Life and Land, which opened in 2004. The Titanis material housed in the FLMNH col- lections is due largely to the efforts of David Webb, his students, and his staff over his 30-year tenure at the FLMNH. It is rumored that the discovery of Titanis precipitated Dave’s life-long pursuit of understanding and documenting the Great American Interchange that is witnessed in the Florida fossil record (Morgan pers comm.). We thought it befitting that the mysterious na- ture of Titanis be revisited as tribute to Dave Webb’s exemplary career in paleontology. In this paper, it is our objective to compile all of the Titanis material housed in the FLMNH’s collections into a comprehensive list and describe the more complete specimens. In so doing, we revisit the prevailing hy- pothesis regarding the morphology of its hand, attempt to more precisely estimate the overall size of Titanis, and consider the evidence for Titanis’ survival into the early Pleistocene. MATERIALS AND METHODS As previously mentioned, the only known North Ameri- can Titanis fossils are from Florida and Texas (Table 1). The four sites in Florida are from Santa Fe River IA and IB, Inglis 1A, and Port Charlotte (Fig. 1), all of which are Pliocene in age (Brodkorb 1963; Carr 1981; Chan- dler 1994; Emslie 1998; Hulbert pers com). The Santa Fe River sites are in-place mid-channel deep-water sedi- ments of Blancan age (Brodkorb 1963; Fig. 1), Inglis 1A is a very late Blancan sinkhole deposit (Carr 1981; Emslie 1998; Fig. 1), and the Port Charlotte site is a canal spoil pile (Hulbert pers comm.; Fig. 1), probably similar in age to Inglis 1A. The Texas material was recovered from a gravel pit along the Nueces River in San Patricio County (near Corpus Christi), Texas. The age of the material is esti- mated to be between 5 million and 15,000 years old Figure 1. Distribution of the sites where Titanis material has been discovered (illustration by Ian Breheny, FLMNH). 203 (Baskin 1995). This single toe bone (TMM 43060-115) is housed at the Texas Memorial Museum (TMM) in Austin. SYSTEMATIC PALEONTOLOGY AND SPECIMENS REVIEWED In total, there are 41 known Titanis specimens in North American museums, all but one reside in the FLMNH’s collections (Table 1). The fossils, isolated cranial and postcranial material, are fragmentary and can fit in two, maybe three specimen drawers. Of these specimens, only 6 have been described in the literature (5 of which have been photographed and/or illustrated; Brodkorb 1963; Chandler 1994; Baskin 1995; Emslie 1998; Alvarenga & Höfling 2003), and the 11 Inglis speci- mens were described in Gail Speaker Carr’s disserta- tion (1981). Herein, we review the best-preserved ma- terial (see Table 1) and include photographs of each of them (Appendix 1). For comparative material, we re- lied on the literature and the available phorusrhacid taxa from the collections at the AMNH/Yale University. Since their initial discovery (Ameghino 1887), more than 20 species within the Phorusrhacidae have been named, renamed, and shuffled among three to four fami- lies (Moreno & Mercerata, 1891; Dolgopol de Saez 1927; Sinclair & Pharr 1932; Brodkorb 1967; Cracraft 1968; Tonni 1980; Alvarenga 1985). Despite the murkiness of the interrelationships within this family, it is widely ac- cepted that they form a monophyletic group (Brodkorb 1967; Tonni 1980; Alvarenga & Höfling 2003). The most recent systematic revision (Alvarenga & Höfling 2003) cites the following characters to support phorusrhacid monophyly: (1) large size; (2) laterally compressed skel- etal elements (premaxilla, thorax, and pelvis) giving the impression of slimness from frontal view; (3) a robust premaxilla with a large hooked beak; (4) a robust man- dibular symphysis; (5) large and pervious nostrils with- out a septum; (6) a desmognathous palate; (7) well de- veloped basipterygoid processes; (8) presence of an ar- ticulation facet for the basipshenoid process on the me- dial side of the pterygoid; (9) absence of uncinate pro- cesses of the ribs; (10) the anterior portion of the pubis is not well developed (as seen in Acciptridae); (11) re- duction of the wings and loss of flight; (12) a coracoid with extreme reduction of the procoracoidal and acrocoracoidal processes with a large scapular facet in the form of a grove; (13) a humerus with a prominent medial tuberosity, the proximal portion of the diaphysis strongly curved with a distally prominent processus flexorius; (14) a tarsometatarsus with a triangular shaped hypotarsus in plantar view and the absence of tendon grooves; and (15) strongly curved ungular phalanx. Phorusrhacid specimens are distributed among various institutions across three continents (South America, Europe, and North America), consequently we had to rely on the literature for descriptions of much of the comparative phorusrhacid material and for the di- mensions considered in our allometric analyses (Appen- dix 2). We used the most recent systematic review of the group (Alvarenga & Höfling 2003) for the nomen- clature and classification presented herein (see Table 3). We considered only those measurements that were consistently conserved across the majority of the more complete specimens, and of course, those that corre- spond to the available Titanis material (Table 1; Appen- dix 2). Forty-one measurements were considered (Table 2) on 16 phorusrhacid taxa, for a total of 26 ‘individuals’ considered in the analyses (Table 3; Appendix 2). It should be noted that composites of the following taxa were used: Titanis walleri, Brontornis burmeisteri (Moreno & Mercerat, 1881), and Physiornis fortis (Ameghino, 1885) (Appendix 2). Although this admis- sion may make some readers wince, we would argue that the determinate growth in birds limits variation in body mass across mature adults within a species (albeit the presence of sexual dimorphism), therefore the mar- gin of variability is minimal. We also chose the largest specimens available to reduce disproportionate composite taxa. The South American Cariamidae (seriemas), an extant group of gruiformes, have been believed to be the sister taxon to the phorusrhacids since the late 1890s (Andrews 1899; Dogopol de Saez 1927; Brodkorb 1967; Cracraft 1968; Livezey 1998; Alvarenga & Höfling 2003). This two-taxon family includes Chunga burmeisteri (Harlaub, 1860) (black-legged seriema) and Cariama cristata (Linneaus, 1776) (red-legged seriema). Sur- prisingly, there are only a handful of cariamids in US museum collections, many of which are either incom- plete or zoo specimens (Appendix 2), and very little has been published on this group. We included the 7 most complete specimens housed at the Smithsonian (USNM) and the AMNH in our analysis (Appendix 2). ESTIMATING THE SIZE OF TITANIS One of the most interesting and relevant questions in biology concerns body size (biomass and body dimen- sions) because it provides useful data to anatomists, ecolo- gists, and biologists (Peters 1983; Schmidt-Nielsen 1984). GOULD and QUITMYER: Titanis walleri: Bones of Contention 204 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb Element Fig. App. 1 Cat. number Locality Date Collected Publication Frontal (partial) UF 137195 SF 1A 1966 Pterygoid (left; cast) Fig. 1 UF 162749 I 1A 1995 (donated) Narial opening, lower margin Fig. 2 UF 137193 SF 1B ? Quadrate (right), orbital process UF 137838 SF 1A 1963 Quadratojugal (left), partial Fig. 3 UF 57580 SF 1A 1963 Chandler,1994 Quadratojugal (right), partial Fig. 3 UF 57585 SF 1A 1966 Chandler, 1994 Mandible (right), articular UF 144179 SF 1B 1993 Vertebra, 2nd cervical (axis) Fig. 4 UF 30006 I 1A 1969, 1974 Carr, 1981 Vertebra, 3rd cervical Fig. 5 UF 30005 I 1A 1969, 1974 Carr, 1981 Vertebra, (partial) UF 30004 I 1A 1969, 1974 Carr, 1981 Vertebra, thorasic (partial) Fig. 6 UF 10415 SF 1A 1965 Coracoid, (partial) UF 10703 SF 1A 1965 Coracoid (left), glenoid fossa UF 144177 SF 1B 1994 Humerus (left), proximal end p/I Fig. 7 UF 137839 SF 1B 1993 Chandler, 1994 Humerus, distal end UF 10418 SF 1A 1963 Carpometacarpus (left) p/I Fig. 8 UF 30003 I 1A 1969, 1974 Carr, 1981; Chandler, 1994; Alvarenga and Hofling, 2003 Femur, shaft UF 12207 SF 1B 1966 Femur, shaft UF 144181 SF 1B 1966 Femur, shaft UF 144182 SF 1B 1966 Fibula, (left) proximal end Fig. 9 UF 9051 SF 1A 1963 Fibula (right), proximal end Fig. 9 UF 7421 SF 1A 1960-1961 Limb, shaft UF 144180 SF 1B 1966 Tibiotarsus (right), proximal end Fig. 10 UF 7333 SF 1A 1963 Tibiotarsus (right) UF 30002 I 1A 1969, 1974 Carr, 1981 Tibiotarsus, shaft UF 137196 SF 1A 1965 Tibiotarsus (left), shaft UF 12208 SF 1B 1966 Tarsometarsis (right), distal end p Fig. 11 UF 4108 (type) SF 1961-1962 Brodkorb, 1963; Alvarenga and Hofling, 2003 Tarsometarsis (right), proximal shaft, medial half UF 137194 SF 1A ? Metatarsal (right), I, partial UF 30007 I 1A 1969, 1974 Carr, 1981 Phalanx UF 10416 SF 1A 1965 Digit III (left), phalanx 1 Fig. 12 UF 30001 I 1A 1969, 1974 Carr, 1981 Digit III (left), phalanx 1 p UF 4109 SF 1961-1962 Brodkorb, 1963 Digit III (right), phalanx 1 Fig. 12 UF 171382 SF 1B 1962 Carr, 1981 Digit III (left), phalanx 2 Fig. 13 UF 30010 I 1A 1969, 1974 Carr, 1981 Digit III (left), phalanx 3 UF 30011 I 1A 1969, 1974 Carr, 1981 Digit IV (left), phalanx, proximal end UF 124228 PC (donated 1990) Digit IV (left), phalanx 1 Fig. 13 UF 30009 I 1A 1969, 1974 Carr, 1981 Digit IV (right), phalanx 1 UF 30008 I 1A 1969, 1974 Carr, 1981 Digit IV (right), phalanx 1 UF 7332 SF 1A 1963 Phalanx, distal end (claw) Fig. 14 UF 10417 SF 1A 1965 Table 1. A comprehensive list of the 40 known fossil elements of Titanis walleri housed at the FLMNH. SF = Santa Fe localities, 1A and 1B; I = Inglis 1A; PC = Port Charlotte. Please refer to the map in Figure 1. Fig # = corresponds to figures in Appendix 1; p i = photographed/illustrated in a previous publication; App. 1 = Appendix 1. 205 With fossil taxa, these estimates are even more critical for accurately reconstructing animals that are often rep- resented by only a few skeletal fragments. Such is the case with Titanis, which as mentioned, is represented by extremely fragmentary material. Estimating its size consequently, is not a straightforward matter. Of first concern is the algorithm itself: body mass is what is commonly calculated, not the height of the animal. Body mass in birds has been estimated using the least shaft circumference of either the tibiotarsus (Campbell & Tonni 1983), or more preferably, the length of the femur (Campbell & Marcus 1988). In the case of Titanis, neither of these elements is preserved in their entirety (Table 1), nor are they preserved consistently in other phorusrhacid taxa. The most recent attempt at estimating the body mass of other phorusrhacids used the least circumference of the distal end of the femur and tarsometatarus which were then compared to ra- tites and other large birds of known weights (Alvarenga & Höfling 2003). It is not clear however, how the au- thors arrived at these height estimates, no algorithm was provided. In this study, we estimate the size of selected skeletal components of Titanis; the length of the skull, the standing height of the bird, and the femur-tibiotarsus- Order Ralliformes Reichenbach, 1852 Suborder Cariamae Fürbringer, 1888 Family Phorusrhacinae Ameghino, 1889 Subfamily Brontornithinae Moreno and Mercerat, 1891 *Brontornis burmeisteri Morena and Mercerat, 1891 (early-middle Miocene) *Physornis fortis Ameghino, 1895 (middle-late Oligocene) *Paraphysornis brasiliensis Alvarenga, 1982 (late Oligocene-early Miocene) Subfamily Phorusrhacinae Ameghino, 1889 *Phorusrhacos longissimus Ameghino, 1887 (early-middle Miocene) Devincenzia pozzi Kraglievich, 1931 (late Miocene-early Pliocene) *Titanis walleri Brodkorb, 1963 (late Pliocene) Subfamily Patagornithinae Mercerat, 1897 *Patagornis marshi Moreno and Mercerat, 1891 (early Miocene) Andrewsornis abbotti Patterson, 1941 (middle Oligocene) *Andalgalornis steulleti Kraglievich, 1931 (late Miocene-early Pliocene) Subfamily Psilopterinae Dolgopol de Saez, 1927 Psilopterus affinis Ameghino, 1899 (middle—late Oligocene) *Psilopterus bachmanni Moreno and Mercerat, 1891 (middle Miocene) *Psilopterus lemoinei Moreno and Mercerat, 1891 (middle Miocene) Psilopterus colzecus Tonni and Tambussi, 1988 (late Miocene) *Procariama simplex Rovereto, 1914 (late Miocene-late Pliocene) Paleopsilopterus itaboraiensis Alvarenga, 1985 (middle Paleocene) Subfamily Mesembriornithinae Kraglievich, 1932 *Mesembriornis milneedwardsi Moreno, 1889 (late Pliocene) Mesembriornis incertus Rovereto, 1914 (late Miocene-early Pliocene) Family Cariamidae Bonaparte, 1853 *Cariama cariama Linnaeus, 1776 (extant) *Chunga burmeisteri Hartlaub, 1860 (extant) Table 2. Proposed taxonomy for the Phorusrhacidae, after Alvarenga and Höfling, 2003. Those taxa indicated by (*) were included in this study. The taxa that are underlined have preserved wing elements. GOULD and QUITMYER: Titanis walleri: Bones of Contention 2 0 6 C E N O Z O IC V E R T E B R A T E S : P apers to H onor S . D avid W ebb Titanis walleri b a X Y Range of X Independent Variable X Dependent Variable Y N r2 Slope Intercept Measurement Predicted (mm) Measurement (mm) Predicted (mm) (mm) (mm) Q-Jugal (L) Skull (TL) 12 0.70 1.19 0.08 170.0 542.3 42.0 - 101.6 Q-Jugal (H) Skull (TL) 10 0.91 0.42 1.98 23.0 359.9 1.0 - 20.3 Axis (L) Skull (TL) 12 0.86 0.64 1.37 127.0 520.5 8.8 - 114.3 Tibiotarsus distal width Femur (L)+Tibiotarsus (L) +Tarsometatarsus (L) 17 0.85 0.74 1.77 88.9 1631.3 14.3 - 63.0 Tarsometatarsus trochlea (W) Femur (L)+Tibiotarsus (L) +Tarsometatarsus (L) 13 0.70 0.52 2.09 76.0 1184.6 8.0 - 105.0 Tarsometatarsus trochlea (W) Standing Height of the Bird 5 0.75 0.60 2.03 76.0 1444.0 20.0 - 105.0 Tibiotarsus distal width Standing Height of the Bird 8 0.87 0.76 1.80 88.9 1870.0 18.5 - 63.0 Table 3. Allometric constants used to predict the skull length, leg length (femur length + tibiotarsus length + tarsometarsus length), and standing height of the bird Titanis walleri. Formula is Y = aXb where X is the independent variable (measured skeletal element, mm); Y is the dependent variable (e.g., skull length, mm); a is the Y-intercept and b is the slope. Q-Jugal = quadratojugal; L= length; H= height; W= width; TL= total length 207 tarsometatarsus length (leg length) using comparative measurements of other phorusrhacids and seriemas and the least-squares regression model commonly used in biology to predict body size relationships (Huxley 1927; 1932, Peters, 1983). We use this method because most animal body size relations can be accurately predicted by the equation (Y = aXb) (Peters 1968). The technique is regarded as a simple and robust approach in describ- ing body size relations (Peters 1983; Schmidt-Nielsen 1984; Reitz et al. 1987). It is also ideally suited to esti- mating body size relationships of incomplete fragmen- tary fossil and subfossil materials (Reitz et al 1987). It is well known that allometry reflects the regular and orderly change of shape, structure, and or function of size among similarly shaped animals (Huxley 1932; Prang et al. 1979; Peters 1983; Schmidt-Nielson 1984; Reitz et al. 1987). Growth is a nonlinear process through ontogeny, and this allometric relationship is described by a mathematical power function y = aXb (Schmidt-Nielson 1984). This is transformed using the common log in order to produce a straight-line regression. The result- ing formula is log y = a + b(log X) with b as the slope of the line, a the y intercept, x the independent variable (skeletal measurement), and y the dependent variable, size estimate of skeletal elements. Allometric analyses generally require a robust sample size and completeness of data (Peters 1983). In our analysis, the data are limited because of the paucity and fragmentary nature of Titanis and phorusrhacid fos- sils, and our inability to take measurements on the taxa that reside in collections outside of the United States. To further complicate the application of least-squares allometry as the predictive tool is the fact that Titanis is one of the largest of the phorusrhacids, consequently, the measurements of the independent variables (X) of the preserved elements (e.g., quadratojugal length) ex- ceeds the calculated regression line based on the smaller phorusrhacid taxa used in the analysis (Table 4). In such instances the confidence limits deteriorate toward the extremes of the regression line. These caveats leave us with the difficult decision of whether or not to continue with an analysis that is known to be compromised a priori or abandon the at- tempt to estimate the size of Titanis. In light of a full- scale reconstruction of Titanis, an expression of Dave Webb’s legacy in vertebrate paleontology, being built for permanent exhibition at the time we conducted this study, we accepted the vagaries of the fossil record and per- severed in our attempt to more accurately estimate its size. And our only practical option to estimating the body relationships of Titanis is through the use of least squares regression. In Table 3 we present the allometric constants used to estimate the total length (TL) of Titanis’ skull (de- pendent variable Y) from the length (L) and width (W) of the largest quadratojugal (independent variable X) preserved. Since the quadratojugal is one of the few cranial elements preserved in Titanis, we had little other options for skull length predictors. As a secondary analy- sis, we also estimated its total skull length (Y) using the axis length (X). Our working assumption being that since the axis vertebra supports the head, its overall size would be a reflection of the skull size it was supporting. The total length of the leg (femur+tibiotarsus +tar- sometatarsus) (Y) and the standing height of the bird was estimated from the width of the distal tibiotarsus and the width of the tarsometatarsus trochlea (Table 3). Our presumption is that these elements must be large enough to support the mass and movement of the ani- mal (Schmidt-Nielsen 1984). The mechanics are simi- lar to engineering supports for flag poles; a 5-foot flag pole requires a much smaller base than does a 100-foot flag pole. It should be noted that we estimated the width of the tarsometatarsus trochlea of Brontornis burmeisteri (Moreno & Mercerat, 1891), believed to be the largest known phorusrhacid, from the sum of the single measurements for each trochlea of specimen (FM- P13259) (Alvarenga & Höfling 2003) and is most likely underestimated. RESULTS AND DISCUSSION Since our primary intent is to further describe the most complete Titanis specimens which are indicated by an asterix (*) in Table 1, we felt that the descriptions should be closely associated with their photographs. Please refer to Appendix I for full descriptions and images of the selected elements. Although we are reluctant to hypothesize about the phylogenetic relationships of Titanis due to our in- ability to adequately compare it to other phorusrhacids, we nonetheless present a cursory review of the only preserved characters proposed by Alvarenga and Höfling (2003) that support the placement of Titanis within the Phorusrhacidae: its gigantic size, the morphology of the pterygoid and tarsometatarsus, and the loss of flight. TITANIS, A PHORUSRHACID? A PHORUSRHACINE? There is a beautifully preserved left pterygoid (UF GOULD and QUITMYER: Titanis walleri: Bones of Contention 208 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb 162749) from the Inglis 1A site, which is represented by a cast (the original resides in a private collection; Ap- pendix 1, Fig. 1). It is 75 mm in length and 15 mm in width with a medially placed facet for articulation of the basipterygoid process (Appendix 1, Fig.1b). The ex- traordinarily large size of the pterygoid in conjunction with the presence of a medially placed articulation facet, are consistent with the characters cited for the family (Alvarenga & Höfling 2003). The distal portion of the tarsometatarsus of Titanis (Appendix 1, Fig. 13c), when viewed from a plantar perspective, is triangular in shape, another cited synapomorphy for the Phorusrhacidae (Alvarenga & Höfling 2003). Alvarenga and Höfling (2003) placed Titanis within the Phorusrhacinae, together with Phorusrhacos long- issimus (Ameghino, 1887) and Devincenzia pozzi (Kraglievich, 1932). The characters they submitted as evidence of monophyly are; a [relatively] long mandibu- lar symphysis that is twice as long as the width of its base, and a [relatively] long and narrow tarsometatar- sus that is approximately 60% of the length of the tibiotarsus. Only fragments of the leg are preserved in Titanis, consisting of only the proximal or distal ends of the tarsometatarus and tibiotarsus (Table 1; Appendix 1, Figs. 13-16). We submit that the placement of Titanis within this subfamily is tenuous until further fossils are recovered (see Table 3). A BIG BIRD The length of the largest of the two preserved quadratojugals (UF 57585) measures 170 mm and esti- mates a total skull length of 542 mm (r2 = 0.70) (Table 3). The total height of quadratojugal (UF 57580) is 23 mm and predicts a skull length of 360 mm (r2 = 0.91) (Table 3). It should be noted that the predicted skull length from the length of the quadratojugal is underestimated because the most distal portion is missing (Appendix 1, Fig. 3). We believe that approximately 25 mm of the bone was not preserved. Regardless, the r2 value for this predictor is not as strong as the height of the quadratojugal, so the missing portion might be a moot point. The axis maximum length of UF 30006 is 127 Figure 2. Graph depicting wing and hind leg lengths for sev- eral phorusrhacid taxa (after Alvarenga, and Höfling, 2003). Figure 3. Photo of the reconstructed foot of Titanis, the carpometacarpus, the distal portion of the humerus, and the carpometacarpus of a wild turkey included for scale (photo by Tammy Johnson, FLMNH). 209 mm, and predicts a total skull length of 521 mm (r2 = 0.86) (Table 3). The allometric predictions of the three equations seem to show that the skull length of Titanis lays somewhere between approximately 360 mm and 542 mm. The width of tarsometatarsus trochlea (UF 4108) is 76 mm and predicts a total leg length (femur+tibiotarsus+tarsometatarsus) of 1185 mm (r2 = 0.70), while the predicted leg length of the distal tibiotarsus width (88.9 mm) yields a value of 1631 mm (r2 = 0.85) (Table 3). The standing height of the bird is predicted to be 1444 mm (r2 = 0.75) based on tarsometarsus trohclea width. The distal width of the tibiotarsus predicts a stand- ing height of 1870 mm (r2 = 0.87). Previously, Titanis had been described as “similar in size to Phorusrhacos longissimus, although differing in proportions … smaller than Devincenzia…” (Brodkorb 1963:115) and between two and three meters tall (Marshall 1994; Feduccia 1999). Our estimate of standing height ranges between 1444 mm and 1870 mm or over 1.5 meters with respect to previous hypotheses of its height. Alvarenga and Höfling (2003) offered estimates of the height of other phorusrhacids, standing height to the top of the back (SHB), and the maximum standing height to the top of the head (MSH), as well as their estimated weights in kilograms: Psilopterus lemoinei (Moreno & Mercerat, 1891) ~ 60cm SHB/80cm MSH, 5 kg P. bachmanni (Moreno & Mercerat, 1891) ~ 60cm SHB/70cm MSH, 5 kg Procariama simplex (Rovereto, 1914) ~ 70cm SHB, 10 kg Patagonis marshi (Moreno & Mercerat, 1891) ~ 90cm SHB, 45 kg Andalgalornis steulleti (Patterson & Kraglievich, 1960) ~ 100cm SHB, 50 kg Mesembriornis milneedwardsi (Moreno, 1889) ~ 110cm SHB/170 cm MSH, 70 kg Phorusrhacos longissimus (Kraglievich, 1931) ~ 130cm SHB/2.4m MSH, 130 kg Paraphysiornis brasiliensis (Alvarenga, 1982) ~ 140cm SHB/240cm MSH, 130 kg Brontornis brumeisteri ~ 175cm SHB/280cm MSH, 350 kg The sister taxa, the seriemas are much smaller, Cariama cariama (Linnaeus, 1776) stands approxi- mately 90 cm at the head, reaching weights of 1.5 kg. Chunga (Hartlaub, 1860) is even smaller, standing at only 50-70 cm at the head and weighing less than 1 kg (Gonzaga, 1996). Based on the estimates of Alvarenga and Höfling (2003), Titanis was one of the taller phorusrhacids, and was probably similar in the size to M. milneedwardsi and A. steulleti, although as we mentioned, our esti- mate is most likely low, hence Titanis could have reached the dimensions of P. longissimus. For relative comparisons, the largest living terres- trial bird, the ostrich (Struthio camelus Linneaus 1758) reaches heights of 200 cm and weighs approximately 130 kg. The largest flying bird known, Argentavis magnificens is estimated to have been weighted be- tween 72 and 79 kg (Campbell & Tonni 1983; Campbell & Marcus 1998). Although the ostrich and Brontornis were similar in height, Brontornis was considerably stockier than the living ostrich. Certainly, much of the weight in phorusrhacids was concentrated in their skulls, which were massive, while the ostrich skull is some- what puny compared to its overall size. Due to the limitations of the preserved specimens, GOULD and QUITMYER: Titanis walleri: Bones of Contention Figure 4. Images of carpometacarpii (in palmar view) of, from right to left, Titanis walleri, (b) Meleagris gallopavo (turkey), (c) Gallus gallus (chicken); photo by Tammy Johnson, FLMNH). 210 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb we were unable to estimate the body mass of Titanis using any of the possible body size relationships (Ap- pendix 2). This remains a an intriguing question for fu- ture research Our estimates for the size of Titanis are as rigor- ous as feasibly possible given the lack of preserved speci- mens. We caution readers not to lose sight of the fact that our analyses were compromised by a small sample size and incomplete of data. A UNIQUE WING? As mentioned, one of the characters that support phorusrhacid monophyly is the loss of flight and reduc- tion of the wing. A review of the maximum lengths of phorusrhacid appendages (i.e., legs and wings) suggests that there is an inverse relationship between the overall size of the bird and the size of its wing (Fig. 2, taken from Alvarenga & Höfling 2003, Fig. 3; Appendix 2). A comparison of the left carpometacarpus, the proximal end of the left humerus, and a reconstruction of the right foot of Titanis illustrate this phenomenon (Fig. 3). The scale used in this image is that of the carpometacarpus of a wild turkey (Meleagris gallopavus Linneaus 1758), their carpometacarpii are approximately the same size! The published descriptions of Titanis’ wing de- scribe it as being a “strong, robust wing (unlike the pae- domorphic wings of ratites) … like its closest relatives … with a rigid wrist and flexible fingers … on a manus that was held extended, as in penguins” (Chandler 1994:176). We agree that Titanis retained a rigid wrist, which is a plesiomorphic state in birds (the ulnare and radiale restrict the movement of the manus in all birds in order to keep primary feathers in alignment [Vasquez 1992]). With respect its robustness, our data suggest otherwise. Figure 2 illustrates the ratio of known phorusrhacid wings to their total leg length; the wing of Titanis is the smallest known with respect to its body size, being approximately 6 times smaller than its leg length, whereas Psilopeterus has a wing length that is only 3 times smaller than its leg length. The hypothesis that the hand of Titanis was “held extended, as in penguins” is purportedly evidenced by an “almost vertical carpal trochlea of the carpometacarpus” (Chandler 1994:176). The carpometacarpus is directed postero-laterally in (most) birds, with the alula pointed downward. Necessarily, the facet on which the carpometacarpus articulates with the ulnare and radiale is vertically oriented. We are unclear as to what the author meant by “almost verti- cal” because we found that in most birds, the orientation of the articulation facet is very similar in morphology, this state is even noted in chicken and turkey wings (Fig. 4; Appendix 1, Fig. 9; see also Gilbert et al. 1981:figs. 123-146). The “presence of a ball joint on the facet of the metacarpal I” instead of an actual (pollix) facet is of- fered as evidence of a flexible claw on the hand of Titanis and its close relatives (Chandler 1994:176). The most recent review of the morphology of the carpometacarpii of phorusrhacids (Alvarenga & Höfling 2003) indicates that the only well preserved phorusrhacid wings are from Titanis, Paraphysiornis, Patagornis, Psilopterus aus- tralis, and Mesembriornis (Alvarenga & Höfling 2003; Table 3). All of them exhibit a protuberance, or ‘ball and joint articulation on the carpometacarpus’, as sug- gested by Chandler (1994). Within most birds, this joint articulates with digit I, or the alula. The first digit in a bird’s hand, although seemingly insignificant, is actually critical in preventing stalling during low-speed flight. The alula, which consists only of phalanx 1, moves indepen- dently of the rest of wing and acts as a wing slot to increase lift. Essentially, the alula and associated feath- ers direct airflow over the upper surface of the wing at a steep angle. As seen in the comparative presentation of carpometacarpii across 64 taxa (Gilbert et al. 1981:figs. 52-59), the morphology of the pollical facet varies interspecifically and can be represented by either a facet or some kind of protruding articulating surface. In fact, a facet for the alula is more common among birds than the retention of a ball joint as seen in Titanis. Presum- ably, there is a relationship between the morphology of the wing and how the bird makes a living, a question that is outside the scope of this study. We do know, however that this ball-joint morphology is also exhibited in seriemas (Alvarenga & Höfling 2003), the closest-living taxon to Phorusrhacidae (see also Gilbert et al. 1981). Neither of the two living seriemas express a flexible claw in lieu of the small single-phalanx finger. Our review of the existing Titanis material did not recover evidence to sug- gest that the expressed phenotype of Titanis and other phorusrhacids was vastly different than that seen in seriemas today (Fig. 5). Chandler (1994) suggested that previous recon- structions of phorusrhacid wings were based on the smaller psilopterine taxa, hence the misinterpretation of wing structure in Titanis and other phorusrhacines. We would argue that, based on the known wing elements (Table 3; Appendix 2) for this group, there is a consider- able range in the ball joint morphotype, suggesting ho- 211 moplastic behavior of this particular character within the Phorusrhacidae. STRATIGRAPHIC RANGE Paleontologists, enthusiasts, and rock companies alike have been actively prospecting and excavating the Florida fossil record for 80+ years. To date, there are approxi- mately 460 documented Pleistocene deposits in Florida (FLMNH sites only) and approximately 86,000 Pleis- tocene vertebrate specimens currently catalogued in the collections at the FLMNH (+/- 2,500 uncatalogued; Hulbert, pers comm). Despite the magnitude of the Pleis- tocene collections, there is no evidence of Titanis in the Florida Pleistocene. Titanis is incredibly rare in the abun- dant Pliocene collections in Florida, which is not unex- pected because large predatory animals are far less abun- dant in a given ecosystem. The Texas site, however, is not as clear-cut. It is reported to be “anywhere from late Hemphillian to late Rancholabrean in age” (Baskin 1995:843) containing early Pliocene horses as well as late Pleistocene verte- brates (Baskin 1995). “The source head for these Hemphillian deposits are unknown but probably eroded from older, undip sediments from the Upper Goliad For- mation and then transported …” and redeposited (Baskin 1995:842). Based on a lack of “definitive Blancan or Irvingtonian taxa” from the site and similarity in color and preservation of the Pleistocene fauna, it was sug- gested that Titanis might have survived into the Rancholabrean (Baskin 1995). Presuming this to be true, it would imply that Titanis was roaming the grasslands of the North American Gulf Coast for more than a mil- lion years since its first known appearance in the fossil record, anywhere. Its location in Texas, and its pre- sumed Rancholabrean age would then suggest that Titanis was heading south. If that were the case, one would still expect to find Titanis in the Florida Pleis- tocene record because, like many other large animals, it would have sought a tropical haven during the ice age. Given the extensive Florida Pleistocene collections, one would expect to find at least some evidence of its exist- ence. An alternative hypothesis is that the Titanis mate- rial in Texas is late Pliocene in age, but the formation from which it came has not yet been identified nor aged appropriately. We would expect to see Titanis in the Pliocene in Texas because certainly, as a migrant in the Great American Interchange, it would have had to pass through Texas to reach Florida; at that time it was a 1,100-mile journey along the Gulf Coast. Unfortunately there are few Pliocene-Pleistocene Gulf sites outside of Florida (Baskin 1995). As depicted in the photographs of the Florida speci- mens in Appendix 1, there is a considerable range in the preservation color of the Titanis elements even in from a single site, as with most Florida Pliocene material. Given the rapid and extreme changes in global sea level during that time (Hulbert 2001), one would expect great perturbations in local Gulf environments coinciding with these sea level changes. Based on these data, we ques- tion the use of color as an indicator for geological age. The only definitive evidence for Titanis is in the Blancan/Irvingtonian of Florida. We reject the use of negative fossil data to support a Pleistocene survival hy- pothesis of Titanis in Texas (i.e., “lack of definitive Blancan taxa” [Baskin 1995]). We prefer the scenario GOULD and QUITMYER: Titanis walleri: Bones of Contention A B Figure 5. (a) A sculpture of Titanis at the Florida Museum of Natural History, (b) A reconstruction of Titanis after Gould & Quitmyer (artwork by Carl Buell, 2002) 212 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb of a yet-to-be discovered Blancan-aged site in Texas in which Titanis bones were deposited, reworked, and re- deposited somewhere down stream. This hypothesis is the most parsimonious given the available data because it does not beg the question as to why Titanis is absent in the Pleistocene record in Florida or why it is not in the Blancan in Texas. And it does further corroborate the faunal interchange hypothesis (Stehli and Webb 1985). CONCLUSIONS Our least squares regression estimates suggest that Titanis stood over 1.5 meters tall (range = 1.4 m – 1.9 m) tall with a skull estimated to have been between 359 mm and 560 mm in length that had a proportionately large beak. As with other phorusrhacids, Titanis most likely had a laterally-compressed physique, long ‘run- ning’ legs and diminutive wings without a claw. The localities in which Titanis has been recovered suggest that it lived in a fairly open grassland environment in which karst sinks and springs were present very much like Florida today. From these sites, a diverse array of taxa have been recovered (Olsen 1965; Scott & Allman 1992; Emslie 1998; Hulbert 2001), to include Xenosmilus hodsonae (sabertooth cat), Eremotherium eomigrans (giant sloth), Glyptotherium arizonae (glyptodont), Rhyncotherium praecursor (proboscidean), Chasmoporthetes ossifragus (hyena), Arctodus pristinus (bear), Erethizon kleini (porcupine), rails, ducks, condors, and other small birds, rodents, lizards, snakes, alligators, turtles, and arthropods. This type of environment is not much different than the environment inhabited by seriemas today. We sus- pect that Titanis, like the seriemas, was an opportunistic feeder that preyed on anything that it could run down and subdue. We also believe that the prey of choice for Titanis was probably anything that it could swallow in its entirety, much like behavior of the seriemas (Gonzaga 1996). As with many other Florida Pliocene taxa, there is no compelling evidence to date that Titanis survived into the Pleistocene, and the Texas evidence is suspect. ACKNOWLEDGEMENTS We gratefully acknowledge the following institutions for providing access to their vertebrate paleontology and ornithology collections: FLMNH, AMNH, and USNM. Specifically, we thank Richard Hulbert for making FLMNH collections available to us, as well as his recol- lections, and Carl Mehling and Mark Norell of the AMNH for allowing access to the phorusrhacid material to us and our artists. We are also very grateful to Bruce MacFadden and the FLMNH’s Exhibits and Public Pro- grams for subsidizing much of this research. A number of people were instrumental in the formation of this work; Ian Breheny for helping us to lay out the images in the body of the text, Tammy Johnson and Jeff Gage of the FLMNH for taking wonderful photographs of Titanis and Cariama, Tara Odorizzi for Photoshop work with the figures, Kurt Auffenberg for helping with the interli- brary loans, Richard Webber who did the skeletal re- construction of Titanis for the FLMNH’s new fossil hall exhibit, Carl D. Buell for the illustration reconstructions of Titanis for this publication and the exhibit, Steve and Suzan Hutchens for reconstructing Titanis’ foot and donating the pterygoid (UF 164729), and Steve Martin of Natural Encounters for allowing us to take images of his seriemas. We also thank Julia Clarke for early dis- cussions of phorusrhacid systematics and morphology, and a rigorous review of this article. And if it were not for the help of Herculano Alvarenga and his disserta- tion, this work would have been greatly impaired. We also thank Luis Chiappe for a thoughtful review of our initial manuscript. We gratefully acknowledge the late Ben Waller who discovered the first Titanis fossil re- mains and brought them to the attention of the Florida Museum of Natural History. Continued paleontological research by Pierce Brodkorb, Steve Emslie, Robert Chandler, Dave Webb and others have kept the research of this most interesting bird alive. Dick Franz, Gary Morgan, Richard Hulbert, John Baskin, and Bruce MacFadden are to be commended for their efforts in putting together this volume. LITERATURE CITED de Alvarenga, H. 1985. Um novo Psilopteridae (Aves: Gruiformes) dos sedimentos terciários de Itaboraí, Rio de Janeiro, Brasil. Congresso Brasileiro de Paleontologia, 8. Série Geologica 27:17-20. de Alvarenga, H. 1999. Revisão sistemática das aves Phorusrhacidae. Ph.D. Dissertation, Instituto de Biociências da Universidade de São Paulo, Brasil. de Alvarenga, H. & E. Höfling. 2003. Systematic revision of the Phorusrhacidae (Aves: Ralliformes). Papeis Avulsos de Zoologia, 43(4):55-91. Ameghino, F. 1887. Enumeración sistemática de las especies de mamíferos fósiles coleccionados por Carlos Ameghino en los terrenos Eocenos de la Patagonia austral y depositados en el Museo de La Plata. Bolltin del Museo de La Plata, 1:1-26. Andrews, C. W. 1899. On the extinct birds of Patagonia. I. 213 The skull and skeleton of Phororhacos inflatus Ameghino. Transactions of the Zoolological Society of London, 15(3):55-86. Baskin, J. A. 1995. The giant flightless bird, Titanis walleri (Aves: Phorusrhacidae) from the Pleistocene coastal plain of south Texas. Journal of Vertebrate Paleontol- ogy, 15(4):842-844. Brodkorb, P. 1963. A giant flightless bird from the Pleistocene of Florida. Auk, 80(2):111-115. Brodkorb, P. 1967. Catalogue of fossil birds, part 3 (Ralliformes, Ichthyornithiformes, Charadriiformes). Bulletin of the Florida State Museum (Biological Sciences), 2(3):99-220. Carr, G. E. 1981. An early Pleistocene avifauna from Inglis, Florida. Ph.D. Dissertation, University of Florida, Gainesville. Campbell, K. E., Jr., & E. P. Tonni 1981. Preliminary observa- tions on the paleobiology and evolution of teratorns (Aves: Teratornithidae). Journal of Vertebrate Paleon- tology, 1:265-272. Campbell, K. E., Jr., & L. Marcus. 1988. The relationships of hindlimb bone dimensions to body weight in birds. Pp. 396-411 in K. E. Campbell, Jr., ed. Papers in avian pale- ontology honoring Pierce Brodkorp. Science Series of the Natural History Museum of Los Angeles. Chandler, R. M. 1994. The wing of Titanis walleri (Aves: Phorusrhacidae) from the late Blancan of Florida. Bulle- tin of the Florida Museum of Natural History, 36(6):175- 180. Cracraft, J. 1968. A review of the Bathornithidae (Aves, Gruiformes), with remarks on the relationships of the suborder Cariamae. American Museum Novitates, 2326:1-46. Dolgopol de Saez, M. 1927. Las aves corredoras fósiles del Sanacrucense. Anales de la Sociedad Científica Argen- tina, 103:145-164. Emslie, S. D. 1998. Avian community, climate, and sea-level changes in the Plio-Pleistocene of the Florida Peninsula. Ornithological Monographs, 50:1-113. Feduccia, A. 1999. The origin and evolution of birds, 2nd ed. Yale University Press, New Haven, 446 p. Gilbert, B. M., L. D. Martin & H. G. Savage. 1981. Avian Osteology. B. M. Gilbert Publisher Laramie, WY, 252 p. Gonzaga, L. P. 1996. Family Cariamidae (Seriemas). Pp. 234- 239 in J. del Hoyo, A. Elliott, & J. Sargatal, eds. Hand- book of the Birds of the World. Vol. 3. Lynx Edicions, Barcelona Hulbert, R. C., Jr. 2001. The Fossil Vertebrates of Florida. University Press of Florida, Gainesville, 350 p. Huxley, J. S. 1927. On the relation between egg-weight and body size in birds. Zoological Journal of the Linnaean Society, 36:457-466. Huxley, J. S. 1932. Problems of Relative Growth, 2nd ed. New York, Dover. Kraglievich, L. 1932. Una gigantesca ave fósil del Uruguay. Devincenzia gallinali n. gen., n. sp., tipo de una neuva familia, Devincenziidae, del orden Stereornithes. Anales del Museo de Historia Natural de Montevideo, 3(2):323- 353. Kraglievich, L. 1940. Descriptcion de la gran ave Pliocena, Mesembriornis milneedwarsi. Pp 639-666 in Obras de geología y paleontología. Recoilación hecha bajo la dirección “ad honorem”, de Alfredo J. Torcelli, puesta en prensa y terminada por Carlos A. Marelli. Taller de Impresiones Oficiales, La Plata Livezey, B. C. 1998. A phylogenetic analysis of the Gruiformes (Aves) based on morphological characters, with an em- phasis on the rails (Rallidae). Transactions of the Royal Society of London. Series B, Biological Sciences, 353(1378):2077-2151. Marshall, L. G., S. D. Webb, J. Joseph, J. J. Sepkoski Jr., & D. M. Raup. 1982. Mammalian evolution and the Great American Interchange. Science, 215:1351-1357. Marshall, L. G. 1994. The terror birds of South America. Sci- entific American, 1994(2):90-95. Moreno, F. P., and A. Mercerat. 1891. Catálogo de los pájaros fósiles de la República Argentina conservados en el Museo de La Plata. [in Spanish and French]. Anales del Museo de La Plata, Sección de Paleontología, 1:1-71. Olsen, S. J. 1965. Vertebrate fossil localities in Florida. Florida Geological Survey Special Publication, 12:1-28. Prang, H. D., J. F. Anderson, & H. Rahn. 1979. Scaling of skeletal mass in birds and mammals. The American Natu- ralist, 113:103-122. Peters, R. H. 1983. The Ecological Implications of Body Size. Cambridge University Press, New York, 329 p. Reitz, E. S., I. R. Quitmyer, H. S. Hale, S. J. Scudder & E. S. Wing. 1987. Application of allometry to zooarchaeology. American Antiquity, 52(2):304-317. Schmidt-Nielsen, K. 1984. Scaling: Why is Animal Size so Important. Cambridge University Press, Cambridge, 241 p. Scott, T. M. & W. D. Allman. 1992. The Plio-Pleistocene stratigraphy and paleontology of southern Florida. Florida Geological Survey Special Publications, 36:1-94. Sinclair, W. J., & M. S. Farr. 1932. Aves of the Santa Cruz beds. Pp. 157-191 in W. Scott, ed. Reports of the Princeton University Expeditions to Patagonia (1896-1899). Princeton University Press, New Haven. Stehli, F. G., & S. D. Webb. 1985. A kalediscope of plates, fauna and flora dispersals, and sea level changes. Pp. 3- 13 in S.D. Webb & F.G. Stehli, eds. The Great American Biotic Interchange. Topics in Geobiology, Plenum Press, New York. Tonni, E. 1980. The present state of knowledge of the Ceno- zoic birds of Argentina. Contributions in Science Series Natural History Museum Los Angeles County, 330:105- 160. Vasquez, R. J. 1992. Functional osteology of the avian wrist and the evolution of flapping flight. Journal of Morphol- ogy, 211:259-268. GOULD and QUITMYER: Titanis walleri: Bones of Contention 214 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb Webb, S. D. 1985. Faunal interchange between North and South America. Acta Zoological Fennica, 170:177-178. Zimmer, C. 1997. Terror take two. Discover Magazine, 18(6):68- 74. 215 Figure 2. Narial opening, ventral bar (left), UF 137193: (a) lateral view; (b) medial view. The anterior most portion of the ventral bar is preserved. It is 135mm in length, its width (mediolateral) at terminal margin of narial opening is 7mm. Appendix 1. The following are images and descriptions of the more complete specimens of Titanis that reside at the FLMNH (photos by Tammy Johnson, FLMNH). Refer to Table 1 for the discovery and publication history of each element listed and Appendix 2 for selected measurements. Figure 1. Pterygoid (right), UF 162749: (a) anterior view; (b) medial view. Almost complete, missing a small portion of its articulation facet for the quadrate. Length is 75mm and its depth, from the lateral edge to articulation facet for the basisphenoid, is 15mm. The articulation facet for the basisphenoid is medially placed, consistent with other phorusrhacids. GOULD and QUITMYER: Titanis walleri: Bones of Contention A B A B 216 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb Figure 3. Quadratojugals: (Top) UF 57580 (left); (Bottom) UF 57585 (right): (a) lateral view; (b) medial view. Only the proximal ends of each are preserved, which includes the articulation tubercular for the quadrate. The larger specimen (Fig.3.2 UF 57585) has a more pronounced crest that is cranial to the articulation tubercular compared to the smaller specimen (Fig 3.1 UF 57580), which has a deep fossa anterocranial to the tubercular. As mentioned by Chandler (1994), there is a distinct difference is size of these bones, possibly indicating sexual dimorphism because there is no indication that the smaller one is a juvenile. Figure 4. Vertebra, 2nd cervical (axis), UF 30006: (a) anterior view; (b) posterior view; (c) lateral view. The specimen is complete. The anterior projection of the process dorsalis is ventrally projected, the dorsal most crest is narrow and rounded at its dorsal terminus. The facies articularis is directed antero-ventrally at a 30° angle. The processus tranversus has enlarged postero- dorsally directed projections, extending just past the articulation surface. A B A B C 217 Figure 5. Vertebra, 3rd cervical, UF 30005: (a) anterior view; (b) posterior view; (c) lateral view. Complete. As in C2, the dorsal terminus of the process dorsalis is rounded at its terminus, however it is positioned more dorsally then in C2 (Carr 1981). The facies articularis is directed antero-dorsally at almost a 90° angle. The processus tranversus have enlarged posteriorly directed projections, the maximum extension is even with the facies articularis. The articulation surfaces of the processus transversus face anteriorly. The orientation of the vertebral arterial canal is directed anteroventrally-posterodorsally. Figure 6. Vertebra, thorasic, UF 10415: (a) anterior view; (b) posterior view; (c) lateral view. This specimen is incomplete, it is missing the processus transversus. The processus dorsalis is extremely pronounced. The facies articularis is directed dorso- medially with a slight upward orientation. The bone is very spongy, suggesting a young individual. GOULD and QUITMYER: Titanis walleri: Bones of Contention A B C A B C 218 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb Figure 7. Humerus (left), proximal end, UF 137839: (a) anterior view; (b) posterior view. The bone is broken just distal to the sulcus for the ligament transverses. See Chandler (1994) for a complete description. Figure 8. Carpometacarpus (left), UF 30003: (a) anterior view; (b) posterior view. This specimen has a broad trochlea carpalis, without a sulcus, and the absence of the process pisiformis. The process intermatacarpus is distinct and located on the proximal edge of the metacarpus minimus. The process alularis is prominent and distinctly rounded in shape. The articular surface for the digit major is flat and without a distinct sulcus. See also Chandler (1994). A B A B 219 Figure 9. Fibula (left), proximal end, UF 9051 and Fibula (right), proximal end, UF 7421: (a) lateral view; (b) medial view. Both specimens are incomplete. The lateral side of the head of the fibulas exhibit a protuberance located anteriorly and just caudal to the crest. Just anterior to this crest is a pronounced facet that runs cranio-caudally. Figure 10. Tibiotarasus (right), proximal end, UF 7333: (a) anterior view; (b) dorsal view; (c) proximal articular surface. The crista cnemialis lateralis is very pronounced, extending far beyond the shaft of the tibiotarsus. The foramen interosseum proximale is absent. The crista cnemialis cranialis is abbreviated, and does not extend down the shaft. The fossa flexoria is well pronounced. The facies articularis lateralis is directed dorsally, with a caudal terminus that extends laterally. GOULD and QUITMYER: Titanis walleri: Bones of Contention A B A B C 220 CENOZOIC VERTEBRATES: Papers to Honor S. David Webb Figure 11. Tarsometatarsus (right), distal end, UF 4108 (type): (a) anterior view; (b) posterior view; (c) plantar view. As described by Brodkorb (1963). Figure 12. Digit III (left), phalanx 1, UF 30001 and Digit III (right), phalanx 1, UF 171382: (a) anterior view; (b) posterior view. A B A B C 221 Figure13. Digit IV (left), phalanx 1, UF 30009 and Digit III (right), phalanx 2, UF 30010: (a) anterior view; (b) posterior view. Figure 14. Phalanx, distal end (claw), UF 10417. GOULD and QUITMYER: Titanis walleri: Bones of Contention A B A B 2 2 2 C E N O Z O IC V E R T E B R A T E S : P apers to H onor S . D avid W ebb Appendix 2. Selected measurements of the more complete elements preserved among the phorusrhacids. These are the data used in our allometry analyses (see Methods). All measurements are in millimeters (mm), weights are in kg. An asterix (*) that accompanies a specimen number indicates they are composites, additional specimen numbers are footnoted. Abbreviations are as follows: Institutions: AMNH; American Museum of Natural History; BMNH = British Museum of Natural History; DGM = Divisao de Geologia e Mineralogia do Departamento Nacional da Producao Mineral; FLMNH = Florida Museum of Natural History; FM = Field Museum of Natural History; MLP = Museo de La Plata; MACN = Museo Argentino de Ciencias Naturales; MMP= Museo de Mar del Plata; PUM = Princeton University Museums; TMM = Texas Memorial Museum; and USNM = Smithsonian Museum of Natural History. H= height; L = length; W = width; D = diameter; C = cervical vertebrae; CMC = carpometacarpus; CMCProx = proximal portion of carpometacarpal; Dig3.1 = digit III, phalanx I; Dig3.1 = proximal portion of digit III, phalanx I; Fem = femur; FemProx = proximal portion of femur; FemDis = distal portion of femor; Fib = fibula; FibDis = distal portion of fibula; Hum = humerus; HumProx = proximal portion of humerus; HumDist = distal most portion of humerus; Mass = weight; MSH = maximum standing height; T-7 = thorasic vertebra #7; T neck = total length of neck; TMT = tarsometatarsus; TMTProx = proximal portion of tarsometatarsus; TMTDMid = midshaft diameter of tarsometatarsus; TMTTroch = tarsometa- tarsus trochlea; TT = tibiotarsus; TTDis = distal portion of tibiotarsus; Sac = sacrum; SHB = standing height to the back. Measurements taken and corresponding columns: (1) skull length: beak tip to paraoccipital process; (2) skull width: cranium at temporal fossae; (3) foramen magnum diameter; (4) length of beak; (5) length of quadratojugal; (6) width of quadratojugal, center of bone: (7) coracoid length. (8) atlas, maximum length; (9) atlas, maximum height; (10) atlas, maximum width, (11) axis, maximum length; (12) axis, maximum height; (13) axis, maximum width; (14) cervical 3, maximum length; (15) cervical 3, maximum height; (16) cervical 3, maximum width: (17) total neck length; T7: (18) thorasic vertebrae number 7 (T-7), maximum length; (19) T-7, maximum height; (20) T-7, maximum width; (21) humerus, maximum length; (22) humerus, maximum width at proximal end; (23) humerus, maximum width at distal end; (24) ulna, maximum length; (25) carpometacarpus (CMC), maximum length; (26) CMC, maximum width at proximal end; (27) femur, maximum length; (28) femur, maximum width at proximal end; (29) femur, maximum width at distal end; (30) tibiotarsus (TT), maximum length, including cnemial crest; (31) TT, maximum width at distal end; (32) fibula, maximum length; (33) fibula, maximum width at distal end; (34) tarsometatarsus (TMT), maximum length; (35) TMT diameter at its proximal end; (36) TMT diameter from mid-shaft; (37) TMT, width through the trochlea; (38) digit III, phalanx I (Dig3.1), maximum length at distal end; (39) Dig3.1, maximum height at the proximal end; (40) sacrum, maximum length; (41) sacrum, maxiumum height from haemal ridge of sacrum to top of iliac crest; (42) standing height of bird from the bottom of the feet to the top of its back (SHB); (43) maximum standing height of bird to the top of its head (MSH); (44) mass. All measurements are recorded in millimeters, weight is recorded in kilograms. 2 2 3 Taxon Specimen Citation Skull (TL) SubfamilyPhorusrhacinae Titanis walleri *composite see Table 1 Devencenzia pozzi (Onactornis depressus) *composite Cabera, 1939 650.0 Phorusrachos inflatus *BMNH A517 Andrews, 1899 337.0 P. inflatus *BMNH A516 ‘type’ Andrews, 1899 337.0 P. longissimus *composite Alvarenga, 1999; Alvarenga & Höfling, 2003 Subfamily Psilopternae Procariama simplex FM P14525 Alvarenga, 1999; Alvarenga & Höfling, 2003 243.0 Psilopterus bachmanni (Pelecyornis puerredonensis) PUM 15904 Sinclair and Farr, 1932 185.0 P. lemoinei (P. tenuirostris) AMNH 9157 Sinclair and Farr, 1932 200.0 P. lemoinei (P. australis) AMNH 9257 Sinclair and Farr, 1932; Alvarenga, 1999; Alvarenga & Höfling, 2003 198.0 P. lemoinei (P. australis) PUM 15109 Sinclair and Farr, 1932 203.0 P. lemoinei (P. australis) PUM 15402 Sinclair and Farr, 1932; Alvarenga, 1999; Alvarenga & Höfling, 2003 185.5 Subfamily Mesembriornithinae Mesembriornis sp. AMNH 7012 (cast) 376.0 M. milneedwardsi MMP S155 Alvarenga, 1999; Alvarenga & Höfling, 2003 440.0 M. milneedwardsi MACN 5944 Kraglievich,1940 340.0 Subfamily Brontornithinae Brontornis burmeisteri *composite Alvarenga, 1999; Alvarenga & Höfling, 2003 Physiornis fortis *composite Alvarenga, 1999; Alvarenga & Höfling, 2003 Paraphysiornis brasiliensis DGM 1418 Alvarenga, 1999; Alvarenga & Höfling, 2003 Subfamily Patagornithinae Patagornis marshi BMNH A516 Alvarenga, 1999; Alvarenga & Höfling, 2003 337.0 Andalgalornis steulleti FM P14357 Alvarenga, 1999; Alvarenga & Höfling, 2003 385.0 Family Cariamidae Cariama cristata AMNH 8904 104.0 Cariama cristata AMNH 1722 120.0 Cariama cristata AMNH 1392 101.6 Cariama cristata FMNH 106728 100.0 Cariama cristata USNM 612030 106.0 Cariama cristata USNM 555731 105.0 Chunga burmeisteri AMNH 4250 104.8 Devencenzia = MHMN 1892,MLP 37-III-7-8; *P. longissimus= AMNH 9146, MLP 131; *B. burmeisteri = MLP 89, 91, TMTTroch W is estimated; *P. fortis= FM P1-3340; MACN A52-185, A52-188 G O U LD and Q U IT M Y E R : Tita n is w a lle ri: B ones of C ontention 2 2 4 C E N O Z O IC V E R T E B R A T E S : P apers to H onor S . D avid W ebb Taxon Specimen Skull (W) Skull FMH Beak (L) Q-Jugal (L) Q-Jugal (W) Coracoid (L) Atlas (L) SubfamilyPhorusrhacinae Titanis walleri *composite 170.0 23.0 Devencenzia pozzi (Onactornis depressus) *composite 191.0 Phorusrachos inflatus *BMNH A517 70.0 17.0 P. inflatus *BMNH A516 ‘type’ 114.0 17.0 157.0 P. longissimus *composite Subfamily Psilopternae Procariama simplex FM P14525 68.0 Psilopterus bachmanni (Pelecyornis puerredonensis)PUM 15904 26.5 12.0 114.0 65.0 4.5 65.0 7.0 P. lemoinei (P. tenuirostris) AMNH 9157 43.0 10.5 78.5 P. lemoinei (P. australis) AMNH 9257 54.0 12.0 104.5 46.4 P. lemoinei (P. australis) PUM 15109 133.0 96.0 P. lemoinei (P. australis) PUM 15402 58.0 9.5 127.0 70.0 7.0 74.0 Subfamily Mesembriornithinae Mesembriornis sp. AMNH 7012 (cast) 117.0 247.0 101.6 20.3 M. milneedwardsi MMP S155 143.0 M. milneedwardsi MACN 5944 Subfamily Brontornithinae Brontornis burmeisteri *composite Physiornis fortis *composite Paraphysiornis brasiliensis DGM 1418 245.0 Subfamily Patagornithinae Patagornis marshi BMNH A516 120.0 Andalgalornis steulleti FM P14357 140.0 Family Cariamidae Cariama cristata AMNH 8904 37.5 6.0 65.0 47.0 1.5 51.5 4.0 Cariama cristata AMNH 1722 27.8 7.9 70.4 51.4 1.0 57.1 3.9 Cariama cristata AMNH 1392 23.2 7.9 63.4 45.4 1.0 49.7 2.3 Cariama cristata FMNH 106728 37.0 7.5 56.5 45.0 1.5 46.5 4.0 Cariama cristata USNM 612030 26.0 6.8 70.7 52.5 1.7 54.0 4.6 Cariama cristata USNM 555731 27.0 61.3 42.0 1.8 52.7 3.5 Chunga burmeisteri AMNH 4250 23.2 9.5 61.5 51.2 1.0 52.5 Devencenzia = MHMN 1892,MLP 37-III-7-8; *P. longissimus= AMNH 9146, MLP 131; *B. burmeisteri = MLP 89, 91, TMTTroch W is estimated; *P. fortis= FM P1-3340; MACN A52-185, A52-188 2 2 5 Taxon Specimen Atlas (H) Atlas (W) Axis (L) Axis (H) Axis (W) C3 (L) C3 (H) C3 (W) T Neck SubfamilyPhorusrhacinae Titanis walleri *composite 127.0 107.0 65.0 141.0 109.0 74.0 Devencenzia pozzi (Onactornis depressus) *composite Phorusrachos inflatus *BMNH A517 P. inflatus *BMNH A516 ‘type’ P. longissimus *composite Subfamily Psilopternae Procariama simplex FM P14525 Psilopterus bachmanni (Pelecyornis puerredonensis)PUM 15904 13.5 14.0 19.0 28.0 24.0 23.0 P. lemoinei (P. tenuirostris) AMNH 9157 19.9 46.6 38.7 P. lemoinei (P. australis) AMNH 9257 29.9 51.3 42.1 P. lemoinei (P. australis) PUM 15109 P. lemoinei (P. australis) PUM 15402 Subfamily Mesembriornithinae Mesembriornis sp. AMNH 7012 (cast) 83.8 63.5 63.5 50.8 M. milneedwardsi MMP S155 M. milneedwardsi MACN 5944 29.0 23.0 49.0 20.0 46.0 62.0 Subfamily Brontornithinae Brontornis burmeisteri *composite Physiornis fortis *composite Paraphysiornis brasiliensis DGM 1418 Subfamily Patagornithinae Patagornis marshi BMNH A516 Andalgalornis steulleti FM P14357 Family Cariamidae Cariama cristata AMNH 8904 15.0 9.0 19.0 19.0 14.0 17.0 12.5 15.0 Cariama cristata AMNH 1722 9.9 9.0 10.1 13.5 7.8 13.7 13.7 10.4 210.0 Cariama cristata AMNH 1392 9.2 7.2 8.8 13.9 5.3 14.1 11.3 12.9 194.0 Cariama cristata FMNH 106728 9.0 8.0 9.0 13.0 12.5 13.5 11.0 14.0 Cariama cristata USNM 612030 9.2 8.9 14.7 13.5 14.2 17.5 13.0 13.6 216 Cariama cristata USNM 555731 8.7 8.5 10.2 11.0 12.1 16.8 12.8 13.9 180 Chunga burmeisteri AMNH 4250 12.6 9.7 13.7 16.3 10.6 13.6 170 Devencenzia = MHMN 1892,MLP 37-III-7-8; *P. longissimus= AMNH 9146, MLP 131; *B. burmeisteri = MLP 89, 91, TMTTroch W is estimated; *P. fortis= FM P1-3340; MACN A52-185, A52-188 G O U LD and Q U IT M Y E R : Tita n is w a lle ri: B ones of C ontention 2 2 6 C E N O Z O IC V E R T E B R A T E S : P apers to H onor S . D avid W ebb Taxon Specimen T-7 (L) T-7 (H) T-7 (W) Hum (L) HumProx (W) HumDis (W) Ulna (L) CMC (L) SubfamilyPhorusrhacinae Titanis walleri *composite 91.0 215.9 76.0 94.0 Devencenzia pozzi (Onactornis depressus) *composite Phorusrachos inflatus *BMNH A517 P. inflatus *BMNH A516 ‘type’ P. longissimus *composite 76.0 Subfamily Psilopternae Procariama simplex FM P14525 104.0 Psilopterus bachmanni (Pelecyornis puerredonensis)PUM 15904 P. lemoinei (P. tenuirostris) AMNH 9157 25.1 38.7 15.3 P. lemoinei (P. australis) AMNH 9257 25.2 44.9 111.0 26.0 23.0 P. lemoinei (P. australis) PUM 15109 P. lemoinei (P. australis) PUM 15402 103.0 24.4 19.5 79.5 47.5 Subfamily Mesembriornithinae Mesembriornis sp. AMNH 7012 (cast) 49.3 138.0 47.4 M. milneedwardsi MMP S155 M. milneedwardsi MACN 5944 18.0 17.0 9.0 80.0 Subfamily Brontornithinae Brontornis burmeisteri *composite Physiornis fortis *composite Paraphysiornis brasiliensis DGM 1418 195.0 83.0 71.5 Subfamily Patagornithinae Patagornis marshi BMNH A516 76.0 76.0 Andalgalornis steulleti FM P14357 Family Cariamidae Cariama cristata AMNH 8904 118.0 Cariama cristata AMNH 1722 10.6 21.2 8.7 106.6 24.8 9.4 54.5 Cariama cristata AMNH 1392 17.5 19.7 11.8 99.0 22.9 11.6 41.3 Cariama cristata FMNH 106728 84.0 Cariama cristata USNM 612030 17.9 26.2 16.5 110.2 24.4 9.0 Cariama cristata USNM 555731 10.5 24.9 21.2 103.0 27.1 7.9 Chunga burmeisteri AMNH 4250 16.9 19.7 20.5 92.7 22.1 6.7 41.5 Devencenzia = MHMN 1892,MLP 37-III-7-8; *P. longissimus= AMNH 9146, MLP 131; *B. burmeisteri = MLP 89, 91, TMTTroch W is estimated; *P. fortis= FM P1-3340; MACN A52-185, A52-188 2 2 7 Taxon Specimen CMCProx (W) Fem (L) FemProx (W) FemDis (W) TT(L) TTDis (W) SubfamilyPhorusrhacinae Titanis walleri *composite 18.0 88.9 Devencenzia pozzi (Onactornis depressus) *composite Phorusrachos inflatus *BMNH A517 227.0 43.0 375.0 43.0 P. inflatus *BMNH A516 ‘type’ 227.0 59.0 62.0 395.0 43.0 P. longissimus *composite 310.0 36.2 92.0 500.0 62.0 Subfamily Psilopternae Procariama simplex FM P14525 158.0 35.0 36.0 292.0 28.0 Psilopterus bachmanni (Pelecyornis puerredonensis)PUM 15904 118.0 22.5 23.5 199.0 18.5 P. lemoinei (P. tenuirostris) AMNH 9157 138.5 29.5 26.0 238.5 21.0 P. lemoinei (P. australis) AMNH 9257 149.5 32.0 33.0 240.0 22.5 P. lemoinei (P. australis) PUM 15109 P. lemoinei (P. australis) PUM 15402 135.0 31.4 30.0 216.0 23.0 Subfamily Mesembriornithinae Mesembriornis sp. AMNH 7012 (cast) M. milneedwardsi MMP S155 277.0 85.0 87.0 458.0 57.0 M. milneedwardsi MACN 5944 13.0 252.0 76.0 78.0 421.0 52.0 Subfamily Brontornithinae Brontornis burmeisteri *composite 420.0 155.0 750.0 63.0 Physiornis fortis *composite 58.0 148.0 Paraphysiornis brasiliensis DGM 1418 350.0 47.0 126.0 550.0 54.0 Subfamily Patagornithinae Patagornis marshi BMNH A516 227.0 59.0 62.0 395.0 27.0 Andalgalornis steulleti FM P14357 Family Cariamidae Cariama cristata AMNH 8904 91.0 21.0 21.0 222.0 21.0 Cariama cristata AMNH 1722 18.0 88.4 20.4 12.2 230.0 15.6 Cariama cristata AMNH 1392 13.2 79.2 16.3 8.7 193.0 14.3 Cariama cristata FMNH 106728 73.0 18.0 17.5 154.5 14.0 Cariama cristata USNM 612030 91.4 21.5 20.5 235.0 15.3 Cariama cristata USNM 555731 85.5 20.7 19.6 216.0 15.9 Chunga burmeisteri AMNH 4250 93.3 70.9 17.9 17.4 179.0 24.7 Devencenzia = MHMN 1892,MLP 37-III-7-8; *P. longissimus= AMNH 9146, MLP 131; *B. burmeisteri = MLP 89, 91, TMTTroch W is estimated; *P. fortis= FM P1-3340; MACN A52-185, A52-188 G O U LD and Q U IT M Y E R : Tita n is w a lle ri: B ones of C ontention 2 2 8 C E N O Z O IC V E R T E B R A T E S : P apers to H onor S . D avid W ebb Taxon Specimen Fib (L) FibDis (W) TMT(L) TMTProx (D) TMTDMid TMTTroch (W) SubfamilyPhorusrhacinae Titanis walleri *composite 91.4 76.0 Devencenzia pozzi (Onactornis depressus) *composite 400.0 110.0 Phorusrachos inflatus *BMNH A517 P. inflatus *BMNH A516 ‘type’ P. longissimus *composite 385.0 80.0 37.0 Subfamily Psilopternae Procariama simplex FM P14525 216.0 30.0 14.0 29.1 Psilopterus bachmanni (Pelecyornis puerredonensis) PUM 15904 16.0 145.0 20.0 9.0 20.0 P. lemoinei (P. tenuirostris) AMNH 9157 179.5 24.5 P. lemoinei (P. australis) AMNH 9257 178.5 26.0 12.0 26.5 P. lemoinei (P. australis) PUM 15109 P. lemoinei (P. australis) PUM 15402 20.0 164.0 10.5 23.7 Subfamily Mesembriornithinae Mesembriornis sp. AMNH 7012 (cast) M. milneedwardsi MMP S155 375.0 61.0 25.0 M. milneedwardsi MACN 5944 360.0 45.0 55.0 Subfamily Brontornithinae Brontornis burmeisteri *composite 400.0 132.0 74.0 *105.0 Physiornis fortis *composite 105.0 67.0 Paraphysiornis brasiliensis DGM 1418 315.0 71.0 Subfamily Patagornithinae Patagornis marshi BMNH A516 280.0 47.0 Andalgalornis steulleti FM P14357 Family Cariamidae Cariama cristata AMNH 8904 200.0 17.0 201.5 18.5 8.0 Cariama cristata AMNH 1722 77.9 11.5 195.0 3.1 4.7 15.3 Cariama cristata AMNH 1392 83.4 10.6 170.0 1.8 3.6 14.0 Cariama cristata FMNH 106728 134.0 10.0 6.5 13.0 Cariama cristata USNM 612030 90.0 13.4 207.0 7.7 14.8 Cariama cristata USNM 555731 83.5 11.7 186.0 9.7 15.4 Chunga burmeisteri AMNH 4250 na na 151.0 2.0 4.7 14.2 Devencenzia = MHMN 1892,MLP 37-III-7-8; *P. longissimus= AMNH 9146, MLP 131; *B. burmeisteri = MLP 89, 91, TMTTroch W is estimated; *P. fortis= FM P1-3340; MACN A52-185, A52-188 2 2 9 Taxon Specimen Dig3.1 (L) Dig3.1Prox (H) Sac (L) Sac (H) *SHB *MSH *Mass SubfamilyPhorusrhacinae Titanis walleri *composite 104.0 48.7 Devencenzia pozzi (Onactornis depressus) *composite 125.0 57.0 Phorusrachos inflatus *BMNH A517 P. inflatus *BMNH A516 ‘type’ 415.0 103.0 P. longissimus *composite 1,300 2,400 130 Subfamily Psilopternae Procariama simplex FM P14525 700 10 Psilopterus bachmanni (Pelecyornis puerredonensis)PUM 15904 28.5 30.0 147.0 38.0 600 700 5 P. lemoinei (P. tenuirostris) AMNH 9157 36.0 39.8 P. lemoinei (P. australis) AMNH 9257 193.0 47.3 P. lemoinei (P. australis) PUM 15109 P. lemoinei (P. australis) PUM 15402 33.0 152.0 600 800 5 Subfamily Mesembriornithinae Mesembriornis sp. AMNH 7012 (cast) 175.0 70.0 M. milneedwardsi MMP S155 1,100 1,700 70 M. milneedwardsi MACN 5944 Subfamily Brontornithinae Brontornis burmeisteri *composite 1,750 2,800 350 Physiornis fortis *composite Paraphysiornis brasiliensis DGM 1418 1,400 2,400 130 Subfamily Patagornithinae Patagornis marshi BMNH A516 900 45 Andalgalornis steulleti FM P14357 1,000 50 Family Cariamidae Cariama cristata AMNH 8904 96.0 Cariama cristata AMNH 1722 50.3 6.8 121.0 23.1 Cariama cristata AMNH 1392 115.0 17.7 Cariama cristata FMNH 106728 79.0 Cariama cristata USNM 612030 23.4 8.3 126.0 26.8 Cariama cristata USNM 555731 24.3 7.7 114.5 26.8 Chunga burmeisteri AMNH 4250 47.0 7.3 116.0 20.4 Devencenzia = MHMN 1892,MLP 37-III-7-8; *P. longissimus= AMNH 9146, MLP 131; *B. burmeisteri = MLP 89, 91, TMTTroch W is estimated; *P. fortis= FM P1-3340; MACN A52-185, A52-188 G O U LD and Q U IT M Y E R : Tita n is w a lle ri: B ones of C ontention