UF MNH bklt BULLETIN UNIVERSITY OF FLORIDA GAINESVILLE DOVES (COLUMBIDAE) AND CUCKOOS (CUCLIDAE) FROM THE EARLY MIOCENE OF FLORIDA David W. Steadman Vol. 48, No. 1, pp. 1-16 2008 ™ STEADMAN: Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida 1 1Florida Museum of Natural History, Dickinson Hall, PO Box 117800, University of Florida, Gainesville, Florida 32611-7800 Steadman, D.W. 2008. Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida. Florida Museum. Nat. Hist. Bull. 48(1):1- 16. DOVES (COLUMBIDAE) AND CUCKOOS (CUCULIDAE) FROM THE EARLY MIOCENE OF FLORIDA David W. Steadman1 ABSTRACT A new genus, Arenicolumba (Aves: Columbiformes: Columbidae), is proposed for the fossil species Columbina prattae Becker and Brodkorb, 1992. This small dove is known from many hundreds of fossils, representing all major post-cranial skeletal elements, from the early Miocene (ca. 18.5 Ma; Hemingfordian land mammal age) Thomas Farm local fauna, Gilchrist County, Florida. Except for the single humerus of Gerandia from the Aquitanian (early Miocene; ca. 22 Ma) of France, Arenicolumba prattae is the oldest known member of the family Columbidae. The nearest living relatives of Arenicolumba appear to be the African genera Oena and Turtur. Thomasococcyx philohippus, new genus and species (Aves: Cuculiformes: Cuculidae), is also described from the Thomas Farm local fauna. Neotropical ground-dwelling cuckoos in the genera Morococcyx, Tapera, Geococcyx, and Dromococcyx may be the closest living relatives of Thomasococcyx. Along with Cursoricoccyx from the Hemingfordian of Colorado, Thomasococcyx philohippus provides evidence that the subfamily of New World ground-dwelling cuckoos (Neomorphinae) existed by the early Miocene. Key Words: Florida; Thomas Farm; early Miocene fossils; Hemingfordian land mammal age; Columbidae; Cuculidae. TABLE OF CONTENTS Introduction .............................................................................................................2 Materials and Methods............................................................................................2 Systematic Paleontology..................................................................................3 Order Columbiformes Family Columbridae.........................................3 Arenicolumba n. gen.................................................................3 Order Cuculiformes Family Cuculidae.................................................5 Thomascoccyx philohippus n. gen. & sp................................ 5 Cuculidae, gen. uncertain........................................................10 Discussion..............................................................................................................10 Columbiformes................................................................................................ 10 Cuculiformes....................................................................................................13 Acknowledgements................................................................................................ 14 Literature Cited...................................................................................................... 15 2 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 48(1) I compared the fossils with the following modern UF skeletons, supplemented by specimens from the University of Kansas Natural History Museum (KU), and the Louisiana State University Museum of Zoology (LSUMZ) — Columbina passerina UF 40074, 43317, C. talpacoti UF 41837, C. buckleyi UF 41837, C. picui UF 38965, C. minuta UF 38786, C. cruziana LSUMZ 48726, 81201, Claravis pretiosa UF 38787, Scardafella inca UF 40806, Metriopelia melanoptera UF 39346, M. ceciliae UF 38788, Geotrygon montana UF 33846, G. chrysia UF 40006, Zenaida macroura UF 40939, Z. aurita UF 40007, Z. asiatica LSUMZ 89696, Leptotila verreauxi UF 41097, L. jamaicensis UF 40027, and Patagioenas [formerly Columba] leucocephala UF 42212. In addition to these 18 species from the Ameri- cas, I also examined modern skeletons of seven Old World species of Columbidae — Oena capensis UF 38785, Turtur chalcospilos UF 38782, Streptopelia capicola UF 38781, Columba guinea UF 38769, Ducula pistrinaria UF 40186, Geopelia cuneata UF 32379, and Gallicolumba jobiensis UF 41460. For the Cuculidae (sequence and subfamilial clas- sification follow Payne 2005), I examined modern skel- etons of these New World species: Crotophaginae — Guira guira UF 38969, Crotophaga major KU 32390, C. ani UF 40013, 42473, C. sulcirostris UF 33855, 40789; Neomorphinae — Tapera naevia UF 38735, Dromococcyx phasianellus KU 91388, D. pavoninus LSUMZ 101257, Morococcyx erythropygus UF 16796, 26250, 26251, 38737, Geococcyx californianus UF 26238, G. velox UF 38729, Neomorphus rufipennis KU 86604, N. geoffroyi LSUMZ 106946; Cuculinae — Coccycua (Piaya) minuta UF 26232, Piaya cayana UF 40787, P. melanogaster KU 88663, Coccyzus americanus UF 40346, C. minor UF 40029, Coccyzus (Hyetornis) pluvialis UF 26233, and Coccyzus (Saurothera) merlini UF 42489. Skeletons of these Old World species also were examined: Centropodinae — Centropus superciliosus UF 33856; Cuculinae — Zanclostomus (Phaenicophaeus) javanicus UF 42704, Phaenicophaeus (Rhopodytes) viridirostris UF 26237, Clamator jacobinus UF 38730, Eudynamys scolopacea UF 41473, Chrysococcyx lucidus UF 39450, and Cuculus canorus UF 35289. For the Musophagidae, I examined skeletons of Corythaeola cristata UF 38728, Corythaixoides concolor UF 38719, Crinifer piscator UF 38718, Musophaga rossae UF 38727, and Tauraco corythaix UF 38726. Measurements were taken with digital calipers, read to the nearest 0.01 mm. Specimens were exam- ined with a Leica MZ8 stereomicroscope, and photo- INTRODUCTION The Thomas Farm site (NAD 27; 29.86°N, 82.83°W, elev. 11 m) is a highly fossiliferous, sediment-filled sink- hole in Gilchrist County, Florida (Pratt 1990). The rich mammalian fauna at Thomas Farm indicates a Hemingfordian land mammal age (LMA) in the early Miocene Epoch, approximately 18.5 million years old (Hulbert 2001, MacFadden 2001). Avian fossils from Thomas Farm include at least 29 species in 25 genera and 18 families (DWS personal observation), although only seven of these species have been described until now. Five of the described species have been assigned to extinct genera, namely the accipitrids Promilio floridanus, P. epileus, and P. brodkorbi, the cracid Boreortalis laesslei, and the phasianid Rhegminornis calobates (Wetmore 1943, 1958, Brodkorb 1954, Olson & Farrand 1974, Steadman 1980, Becker 1987). An apparent anhinga, originally described as a cormorant Phalacrocorax subvolans by Brodkorb (1956), was assigned to the living genus Anhinga but without details by Becker (1987). The seventh described species from Thomas Farm is a small columbid, Columbina prattae Becker and Brodkorb (1992). This dove is by far the most abundant species of bird at Thomas Farm. Assigning Columbina prattae to an extant genus sets this species apart from the rest of the avifauna, with the exception of Anhinga subvolans. In this paper I re-examine the osteology of C. prattae, and conclude that this dove is like most other birds (and all mammals; Hulbert 2001, MacFadden 2001) from Thomas Farm in being generically distinct from living species. I also describe a new genus and species of cuckoo (Cuculidae), a family previously unreported from Tho- mas Farm and with a limited fossil record in North America. MATERIALS AND METHODS The fossils from Thomas Farm are housed in the Divi- sion of Vertebrate Paleontology, Florida Museum of Natural History, University of Florida. Certain of the fossils, formerly in the private collection of Pierce Brodkorb, have the catalogue prefix UF/PB. Some of the modern skeletons also were originally in the private collection of Pierce Brodkorb; these specimens have a PB catalogue number and a UF catalogue number as- signed in the 1990s. Both the UF and PB numbers are listed for these specimens in the figure legends. UF is used as a prefix for both the fossils housed in the Verte- brate Paleontology Collection and for modern skeletons housed in the Ornithology Collection. STEADMAN: Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida 3 graphed with a Sony DSC-R1 camera. Osteological ter- minology is from Howard (1929) and Baumel et al. (1993). Character polarities were not proposed because of a lack of strong, consistent evidence of which fami- lies would be appropriate as outgroups for the Columbidae or Cuculidae. SYSTEMATIC PALEONTOLOGY Order Columbiformes (Latham 1790) Family Columbidae (Illiger 1811) Six of the seven characters (based on three skel- etal elements) that Becker and Brodkorb (1992) used to assign the species prattae to the genus Columbina are, in fact, found in many living genera of Columbidae world- wide. Their seventh character [“the dorsal surface of trochlea IV extending dorsad (in distal view) past the level of trochlea III”] does distinguish the tarsometatar- sus of Columbina sensu lato (i.e., including Claravis, Scardafella, and Metriopelia) from that of Geotrygon, Zenaida, Leptotila, and Patagioenas but not from that in Old World genera such as Turtur and Ducula. In the diagnosis of the new genus, I will begin with the tarsometatarsus since that is the holotype (UF 106594) of Columbina prattae as well as the skeletal element with the largest number of diagnostic charac- ters. Arenicolumba n. gen. Diagnosis.—A small columbid (Table 1) that dif- fers from other genera (especially Columbina, Claravis, Scardafella, Metriopelia, Geotrygon, Zenaida, Leptotila, and Patagioenas) as follows. Tarsometatar- sus (Fig. 1B, C): 1, in acrotarsial aspect, intercotylar knob with flatter proximal surface (less rounded, less pointed); 2, facet for hallux located more proximad than in all genera except Patagioenas; 3, larger foramen vasculare distale; 4, crista plantare lateralis more dis- tinct than in all genera except Geotrygon; 5, in both plantar and acrotarsial aspect, the articulating surface of trochlea metatarsi III shorter than in all genera ex- cept Patagioenas; 6, in distal aspect, trochlea metatarsi III wide relative to its depth; 7, central groove in tro- chlea metatarsi III deeper. Tibiotarsus (Fig. 2B): 8, inci- Coracoid: Tarsometatarsus: depth of distal width processus Tibiotarsus: (without medio-plantar acrocoracoideus distal width process) Arenicolumba prattae 2.47 3.39 4.09 Thomas Farm fossils 2.08–2.69 3.33-3.42 4.02–4.22 18 3 3 Columbina passerina F 2.05 2.83 3.28 Florida 1.86–2.13 2.65–2.97 3.12–3.50 12 12 12 Columbina passerina M 2.15 2.82 3.35 Florida 1.96–2.28 2.48–3.20 3.17–3.50 14 14 14 Zenaida macroura F 3.11 3.97 4.53 Florida 2.92–3.32 3.80–4.22 4.29–4.77 20 20 20 Zenaida macroura M 3.30 4.08 4.66 Florida 3.07–3.63 3.79–4.27 4.38–4.95 20 20 20 Table 1. Measurements (in mm) of Arenicolumba prattae (Hemingfordian LMA, Thomas Farm, Florida) compared with those of adult skeletons of two widespread, living North American species, Columbina passerina and Zenaida macroura (F = female, M = male). 4 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 48(1) Figure 2. The tibiotarsus of columbids in dorsal (upper row) and ventral (lower row) aspects. A. Columbina passerina UF 40074. B. Arenicolumba prattae UF 96296. C. Oena capensis UF 38785 [PB 39624]. D. Turtur chalcospilos UF 38782 [PB 36021]. Scale = 10 mm. Figure 1. The tarsometatarsus of columbids in acrotarsial (upper row) and plantar (lower row) aspects. A. Columbina passerina UF 40074. B. Arenicolumba prattae UF 62093. C. Arenicolumba prattae UF 219393. D. Oena capensis UF 38785 [PB 39624]. E. Turtur chalcospilos UF 38782 [PB 36021]. Scale = 10 mm. STEADMAN: Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida 5 five and six of these 11 characters, respectively, sug- gests that Arenicolumba may be more closely related to these genera than to any extant New World genus of Columbidae. Today, the single species of Oena and the five species of Turtur are confined to Africa (Gibbs et al. 2001). Order CUCULIFORMES (Wagler 1830) Family CUCULIDAE Vigors 1825 A set of 12 fossils representing five skeletal ele- ments is referred to the Cuculidae because of this com- bination of characters. Tarsometatarsus: two hypotarsal canals; intercotylar knob prominent; in plantar aspect, fused area of cristae hypotarsi short; in proximal aspect, cotyla medialis square; corpus tarsometatarsi relatively uniform in width, and much wider than deep; trochlea lateralis greatly elevated, dorso-ventrally expanded, and antero-posteriorly compressed. Coracoid: facies articularis sternalis developed on both dorsal and ventral surfaces; facies articularis sternalis evenly developed on dorsal surface. Scapula: facies articularis humeralis well elevated from corpus scapulae; acromion blunt in dorsal aspect; corpus scapulae of uniform depth through much of its length. Humerus: distal end greatly expanded ventrad, yielding an elongated processus flexorius and epicondylus ventralis; fossa musculo brachialis distinct and oval. Mandibular ramus: similar in thickness, curva- ture, and overall proportions to that in Tapera and Morococcyx. Thomasococcyx philohippus n. gen. & sp. Holotype.—Proximal end of right tarsometatarsus, UF 96268 (Fig. 6A), Thomas Farm local fauna sura intercondylaris relatively wide (condylus medialis and condylus lateralis relatively narrow). Sternum (Fig. 3B): 9, rostrum sterni (manubrium) more elongate. Co- racoid (Fig. 4B, C): 10, in dorsal aspect, processus procoracoideus extends farther sternally. Humerus (Fig. 5B, C): 11, incisura capitis extends farther dorso-ven- trally to wrap around the ventral margin of tuberculum ventral and separate it more from corpus humeri. Type species.—-Arenicolumba prattae is the type and only known species in the genus. Etymology.—From the Latin arena, a feminine word that means “sand, sandy place” (Brown 1956:678) and the Latin columba, a feminine word meaning “dove, pigeon” (Brown 1956: 278). The second a in arena be- comes a linking i because it looks and sounds better this way (Winston 1999: 163). The name Arenicolumba re- fers to the sandy sediment at Thomas Farm (see Pratt 1990) in which fossils of this dove and other small verte- brates are commonly found. Remarks.—The 11 diagnostic characters of Arenicolumba also distinguish it from the seven Old World genera of Columbidae (see MATERIALS AND METHODS) except for agreement with Oena in char- acters 1, 3, 5, 9, and 10, with Turtur in characters 2, 3, 5-7, and 10, with Streptopelia in character 4, with Columba in characters 4 and 10, with Ducula in char- acters 2 and 4, with Geopelia in characters 4 and 10, and with Gallicolumba in characters 4-6 and 10. Note in the diagnosis that Arenicolumba differs in all 11 diagnostic characters from all four New World genera/subgenera (depending on classification) of small columbids, namely Columbina, Claravis, Scardafella, and Metriopelia. Agreement with Oena and Turtur in Figure 3. The sternum of columbids in ventro-anterior aspect. A. Columbina passerina UF 40074. B. Arenicolumba prattae UF 219359. C. Oena capensis UF 38785 [PB 39624]. D. Turtur chalcospilos UF 38782 [PB 36021]. Scale = 10 mm. 6 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 48(1) Figure 4. The coracoid of columbids in dorsal (upper row) and ventral (lower row) aspects. A. Columbina passerina UF 40074. B. Arenicolumba prattae UF 223384. C. Arenicolumba prattae UF/PB 6503. D. Oena capensis UF 38785 [PB 39624]. E. Turtur chalcospilos UF 38782 [PB 36021]. Scale = 10 mm. Figure 5. The humerus of columbids in palmar (upper row) and anconal (lower row) aspects. A. Columbina passerina UF 40074. B. Arenicolumba prattae UF 223471. C. Arenicolumba prattae UF 223408. D. Oena capensis UF 38785 [PB 39624]. E. Turtur chalcospilos UF 38782[PB 36021]. Scale = 10 mm. STEADMAN: Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida 7 (Hemingfordian LMA), Gilchrist County, Florida. Paratypes.—Distal ends of tarsometatarsi UF 96268, 106406, 106595, 106735, 223532 (Fig. 7A), man- dibular rami UF 219354, 219355, sternal end of coracoid UF 106422 (Fig. 8A), coracoidal ends of scapulae UF 106451, 106638 (Fig. 9A), distal end of humerus UF 223499 (Fig. 10A), all from the Thomas Farm local fauna (Hemingfordian LMA), Gilchrist County, Florida. Diagnosis.—A medium-sized, presumably ground- dwelling cuckoo defined by the following combination of characters. Tarsometatarsus: cotyla lateralis more anconally protrudent than in Cursoricoccyx; foramina hypotarsi relatively larger than in Cursoricoccyx; hypotarsus extends more mediad than in Cursoricoccyx; sulcus extensorius relatively deeper than in Cursoricoccyx, Geococcyx, Neomorphus, Coccyzus, Clamator, Chysococcyx, or Cuculus; sulcus extensorius relatively narrower than in Guira, Crotophaga, Dromococcyx, Geococcyx, Neomorphus, Coccycua, Piaya, Centropus, Phaenicophaeus, Clamator, Eudynamys, Chrysococcyx, or Cuculus; combined depth and width of sulcus extensorius most similar to the condition in Tapera and Morococcyx; corpus tar- sometatarsi narrower relative to its depth than in any other genus, this difference being most pronounced com- pared to specimens of Old World genera; foramen vasculare distale larger than in all other genera. Cora- coid: resembles that in Crotophaga, Morococcyx, Figure 6. The proximal end of the tarsometatarsus of cuculids in acrotarsial (upper row), plantar (middle row), and proximal (lower row) aspects. A. Thomasococcyx philohippus UF 96268 (Neomorphinae). B. Morococcyx erythropygus UF 38737 [PB 29340] (Neomorphinae). C. Tapera naevia UF 38735 [PB 22915] (Neomorphinae). D. Geococcyx velox UF 38729 [PB 17647] (Neomorphinae). E. Crotophaga sulcirostris UF 40789 (Crotophaginae). F. Coccyzus americanus UF 40346 (Cuculinae). G. Clamator jacobinus UF 38730 [PB 27715] (Cuculinae). Scale = 10 mm. 8 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 48(1) Figure 7. The distal end of the tarsometatarsus of cuculids in acrotarsial (upper row) and plantar (lower row) aspects. A. Thomasococcyx philohippus UF 223532 (Neomorphinae). B. Morococcyx erythropygus UF 38737 [PB 29340] (Neomorphinae). C. Tapera naevia UF 38735 [PB 22915] (Neomorphinae). D. Geococcyx velox UF 38729 [PB 17647] (Neomorphinae). E. Crotophaga sulcirostris UF 33855 (Crotophaginae). F. Coccyzus americanus UF 40346 (Cuculinae). G. Clamator jacobinus UF 38730 [PB 27715] (Cuculinae). Scale = 10 mm. Figure 8. The coracoid of cuculids in dorsal (upper row), ventral (middle row), and sternal (lower row) aspects. A. Thomasococcyx philohippus UF 106422 (Neomorphinae). B. Morococcyx erythropygus UF 38737 [PB 29340] (Neomorphinae). C. Tapera naevia UF 38735 [PB 22915] (Neomorphinae). D. Geococcyx velox UF 38729 [PB 17647] (Neomorphinae). E. Crotophaga sulcirostris UF 40789 (Crotophaginae). F. Coccyzus americanus UF 40346 (Cuculinae). G. Clamator jacobinus UF 38730 [PB 27715] (Cuculinae). Scale = 10 mm. STEADMAN: Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida 9 Figure 9. The scapula of cuculids in dorsal aspect. A. Thomasococcyx philohippus UF 106638 (Neomorphinae). B. Cuculidae sp. UF/PB 6487. C. Morococcyx erythropygus UF 38737 [PB 29340] (Neomorphinae). D. Tapera naevia UF 38735 [PB 22915] (Neomorphinae). E. Geococcyx velox UF 38729 [PB 17647] (Neomorphinae). F. Crotophaga sulcirostris UF 40789 (Crotophaginae). G. Coccyzus americanus UF 40346 (Cuculinae). H. Clamator jacobinus UF 38730 [PB 27715] (Cuculinae). Scale = 10 mm. Figure 10. The humerus of cuculids in palmar (upper row) and anconal (lower row) aspects. A. Thomasococcyx philohippus UF 223499 (Neomorphinae). B. Morococcyx erythropygus UF 38737 [PB 29340] (Neomorphinae). C. Tapera naevia UF 38735 [PB 22915] (Neomorphinae). D. Geococcyx velox UF 38729 [PB 17647] (Neomorphinae). E. Crotophaga sulcirostris UF 40789 (Crotophaginae). F. Coccyzus americanus UF 40346 (Cuculinae). G. Clamator jacobinus UF 38730 [PB 27715] (Cuculinae). Scale = 10 mm. 10 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 48(1) two fossils respectively represent 8.5 mm (left) and 11.5 mm (right) lengths of the fused prearticular, articular, and dentary of what is likely to have been a single man- dibular ramus that was similar in size to that of Morococcyx (as well as Tapera and Dromococcyx). Cuculidae, gen. uncertain Material.—Scapula lacking distal end, UF/PB 6487 (Fig. 9). Description.— This scapula compares closely in overall size to that of Guira guira, Crotophaga ani, Coccyzus (Saurothera) merlini, and Piaya cayana. It is much too large to pertain to Thomasococcyx philohippus (Table 2), from which it differs further by having a much more dorsally flared facies articularis clavicularis as in Tapera or Crotophaga. UF/PB 6478 is approximately the right size for Cursoricoccyx gertrudae (Hemingfordian of Colorado; Martin and Mengel 1984), based on extrapolation from the size of the coracoid and carpometecarpus in the latter. More material of this large cuckoo from Thomas Farm is needed before it can be characterized further. DISCUSSION COLUMBIFORMES Compared to many other non-passerine orders of birds, the Columbiformes (traditionally but probably er- roneously considered to consist of Pteroclididae + Columbidae) has a poor fossil record in the Tertiary. The Pteroclididae (sandgrouse) is represented by sev- eral Paleogene (late Eocene – early Oligocene) fossils from Europe that represent two extinct genera with no apparent links to the Columbidae (Archaeoganga, Leptoganga; Mourer-Chauviré 1992, 1993). I see no characters in Arenicolumba that would link it more than any other columbid to the Pteroclididae or any other fam- ily. The Columbidae has no Paleogene fossil evidence at all, even among the very rich European and North American avifaunas of the Eocene and Oligocene, where the first evidence of many if not most living orders and families of non-passerine birds is found (James 2005, Mayr 2005). Nothing in the fossil record of columbids or of birds in general supports the Cretaceous origin of Columbiformes proposed by Pereira et al. (2007) from mitochondrial and nuclear genetic sequences. Because the Columbiformes almost certainly did not exist in the Cretaceous, one need not invoke the breakup of Gondwana as an essential part of their historic biogeog- raphy. Nevertheless, the Columbiformes may have origi- nated in the Southern hemisphere (Olson 1989), an idea supported by their absence in the rich Paleogene fossil Geococcyx, and Neomorphus more than in other gen- era in the medial-lateral placement of, and great dorsal- ventral depth of, the ventral shelf of facies articularis sternalis (and resulting bulge of corpus coracoidei); agrees with that in Tapera, Morococcyx, and Geococcyx, but differs from that in other genera, in having a shallow but sterno-humerally expanded dorsal component of facies articularis sternalis. Scapula: relative to overall size of the bone, facies articularis humeralis smaller than in all other genera, and less elongated than in all New World genera; acromion less elongate than in Guira, Crotophaga, Tapera, Dromococcyx, and Neomorphus; facies lateralis of corpus scapulae, be- tween facies articularis clavicularis and facies articularis humeralis, more concave than in all New World genera except Guira and Neomorphus. Humerus: processus flexorius smaller (less expanded posteriorly) than in all genera except Dromococcyx and Geococcyx. Etymology.—- Thomasococcyx is from Thomas, referring to the Thomas Farm locality, and coccyx (from the Greek kokkyx, m., cuckoo; Brown 1956:216, 241). The Thomas family still cultivates much of the land sur- rounding the fossil site. Thomasococcyx is masculine. philohippus is from philos (Greek, m., dear one, friend; Brown 1956:355) and hippos (Greek, m., horse; Brown 1956:423), an allusion to the three genera and species of three-toed horses (see MacFadden 1992, 2001) that shared the northern Florida landscape with this cursorial cuckoo in the early Miocene. One of these horses, Parahippus leonensis, is particularly common at Tho- mas Farm (Hulbert 1984, 2001). The name philohippus is a masculine noun in apposition. Remarks.— Thomasococcyx philohippus is the type and only known species in the genus. The tarsometa- tarsus of Thomasococcyx is most similar qualitatively to that in Tapera and Morococcyx among living genera of cuckoos, whereas the coracoid and scapula are most similar qualitatively to those in Morococcyx and Geococcyx. The humerus of Thomasococcyx is most similar qualitatively to that in Dromococcyx and Geococcyx. Thus, Thomasococcyx is, as far as can be determined now, a mosaic of living genera of Neomorphinae. The coracoid, scapula, humerus, and tarsometa- tarsus of Thomasococcyx philohippus are all about the size of the same elements in Morococcyx erythropygus (Table 2), and therefore I interpret these four skeletal elements to belong to the same extinct spe- cies. I am somewhat less confident about referral of the two mandibular rami (UF 219354, 219355) to Thomasococcyx, given the great variation in bill depth among living genera of cuckoos. Nevertheless, these STEADMAN: Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida 11 Sc ap ul a: l en gt h Sp ec ie s M an di bl e: d ep th a t C or ac oi d: C or ac oi d: l en gt h an d w id th o f Sc ap ul a: w id th o f Ta rs om et at ar su s: ju nc tio n of d en ta ry a nd st er na l w id th th ro ug h sc ap ul ar fa ci es a rti cu la ris co rp us H um er us : C ar po m et ac ar pu s: Ta rs om et at ar su s: Ta rs om et at ar su s: m in im um d ep th o f c or pu s su ra ng ul ar an d de pt h an d gl en oi d fa ce ts hu m er al is sc ap ul ae di st al w id th pr ox im al w id th pr ox im al w id th pr ox im al d ep th ta rs om et at ar si Th om as oc oc cy x 2. 15 ( U F 21 93 55 ) 4. 03 , 1 .9 0 2. 48 , 1 .9 1 1. 89 6. 30 --- 4. 22 4. 51 1. 14 ph ilo hi pp us , 2. 18 ( U F 21 93 54 ) (U F 10 64 22 ) --- (U F 10 64 51 ) (U F 10 64 51 ) (U F 22 34 99 ) (U F 96 26 8) (U F 96 26 8) (U F 10 64 06 ) ne w s pe ci es 2. 47 , 1 .6 9 1. 67 (U F 10 66 38 ) (U F 10 66 38 ) C ur so ri co cc yx -- - --- 5. 51 --- --- --- 6. 03 5. 68 5. 57 --- ge rt ru da e (K U V P 25 63 0) (K U V P 25 63 1) (K U V P 25 62 9) (K U V P 25 62 9) K U V P 25 62 9- 25 63 1 C uc ul id ae s p. --- --- --- 3. 18 , 2 .2 9 2. 12 --- --- --- --- --- Th om as F ar m U F/ PB 6 48 7 G ui ra g ui ra 3. 58 5. 84 , 2 .6 2 5. 51 3. 49 , 2 .2 8 2. 10 7. 96 5. 86 5. 49 5. 71 1. 48 U F 38 96 9 C ro to ph ag a m aj or 5. 32 7. 64 , 2 .2 0 5. 30 3. 83 , 2 .3 2 2. 26 9. 56 --- 6. 10 6. 73 1. 49 K U 3 23 90 C ro to ph ag a an i 4. 53 5. 71 , 2 .3 8 4. 48 3. 41 , 2 .5 1 1. 95 7. 62 5. 48 5. 58 5. 58 1. 47 U F 42 47 3 C ro to ph ag a 3. 81 4. 91 , 1 .9 4 3. 64 2. 59 , 1 .8 9 1. 67 6. 17 4. 63 4. 21 4. 78 1. 23 su lc ir os tr is U F 40 78 9 Ta pe ra n ae vi a 2. 33 3. 92 , 1 .6 3 3. 37 2. 52 , 1 .4 8 1. 48 5. 76 4. 29 4. 00 3. 84 0. 99 U F 38 73 5 D ro m oc oc cy x 2. 28 5. 91 , 2 .2 3 4. 62 3. 04 , 2 .4 8 2. 00 8. 42 --- 4. 59 5. 06 1. 25 ph as ia ne llu s K U 9 13 88 D ro m oc oc cy x 2. 04 4. 03 , 1 .7 1 --- 2. 50 , 1 .7 1 1. 53 7. 25 --- 3. 98 4. 29 --- pa vo ni nu s LS U M Z 10 12 57 M or oc oc cy x 2. 64 4. 59 , 1 .5 4 3. 55 2. 34 , 1 .8 5 1. 60 5. 71 3. 87 4. 20 4. 14 1. 25 er yt hr op yg us U F 38 73 7 G eo co cc yx 4. 90 8. 78 , 3 .4 9 6. 27 3. 60 , 2 .7 0 2. 50 9. 74 7. 26 8. 24 8. 33 2. 41 ca lif or ni an us U F 26 23 8 Ta bl e 2. M ea su re m en ts (i n m m ) o f T ho m as oc oc cy x ph ilo hi pp us , n ew g en us a nd sp ec ie s ( H em in gf or di an L M A , T ho m as F ar m , F lo rid a) c om pa re d to th os e of Cu rs or ic oc cy x ge rtr ud ae ( H em in gf or di an L M A , M ar tin C an yo n, C ol or ad o) , a n in de te rm in at e cu cu lid f ro m T ho m as F ar m , a nd a du lt sk el et al e le m en ts o f m od er n sp ec ie s o f C uc ul id ae . 12 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 48(1) Ta bl e 2. (C on t.) Sc ap ul a: l en gt h Sp ec ie s M an di bl e: d ep th a t C or ac oi d: C or ac oi d: l en gt h an d w id th o f Sc ap ul a: w id th o f Ta rs om et at ar su s: ju nc tio n of d en ta ry a nd st er na l w id th th ro ug h sc ap ul ar fa ci es a rti cu la ris co rp us H um er us : C ar po m et ac ar pu s: Ta rs om et at ar su s: Ta rs om et at ar su s: m in im um d ep th o f c or pu s su ra ng ul ar an d de pt h an d gl en oi d fa ce ts hu m er al is sc ap ul ae di st al w id th pr ox im al w id th pr ox im al w id th pr ox im al d ep th ta rs om et at ar si G eo co cc yx v el ox 4. 14 7. 32 , 2 .1 8 5. 09 3. 86 , 2 .4 5 2. 40 8. 43 6. 07 7. 07 6. 68 2. 13 U F 38 72 9 N eo m or ph us 4. 77 7. 79 , 2 .3 6 5. 96 4. 07 , 2 .5 4 2. 37 9. 19 --- 8. 44 8. 49 2. 31 ru fip en ni s K U 8 66 04 N eo m or ph us 5. 41 8. 73 , 2 .3 2 --- 4. 67 , 3 .4 0 2. 83 10 .5 4 --- 9. 41 9. 23 2. 62 ge of fr oy i LS U M Z 10 69 46 Co cc yc ua ( Pi ay a) 1. 98 3. 42 , 1 .4 8 3. 14 1. 86 , 1 .5 2 1. 20 5. 38 3. 82 3. 53 3. 50 0. 92 m in ut a U F 26 23 2 Pi ay a ca ya na 3. 32 5. 56 , 1 .5 1 4. 10 2. 93 , 2 .0 8 2. 00 7. 77 5. 76 5. 43 5. 56 1. 43 U F 40 78 7 Pi ay a m el an og as te r 3. 53 6. 50 , 1 .8 0 4. 26 2. 87 , 2 .2 8 2. 13 7. 56 --- 5. 72 5. 23 1. 87 U F 40 78 7 C oc cy zu s am er ic an us 2. 90 4. 66 , 1 .8 2 4. 03 2. 43 , 1 .8 5 1. 63 6. 04 5. 14 4. 25 4. 11 1. 19 U F 40 34 6 C oc cy zu s m in or 3. 00 5. 43 , 2 .0 0 3. 91 2. 46 , 1 .9 0 1. 61 6. 24 5. 10 4. 69 4. 63 1. 18 U F 40 02 9 C oc cy zu s (H ye to rn is ) --- 7. 17 , 2 .3 3 5. 25 3. 79 , 2 .8 1 2. 43 9. 29 --- --- --- --- pl uv ia lis U F 26 23 3 C oc cy zu s (S au ro th er a) 4. 90 5. 96 , 1 .8 1 4. 31 3. 21 , 2 .2 6 2. 23 7. 45 5. 56 6. 40 6. 24 1. 68 m er lin i U F 42 48 9 C en tr op us 4. 02 6. 62 , 3 .2 5 5. 25 3. 84 , 2 .4 3 2. 13 8. 25 6. 41 6. 46 6. 70 2. 00 su pe rc ili os us U F 33 85 6 Za nc lo st om us 3. 08 5. 56 , 1 .7 0 3. 70 2. 87 , 1 .9 4 1. 67 7. 11 5. 26 4. 68 4. 78 1. 19 (P ha en ic op ha eu s) ja va ni cu s U F 42 70 4 C la m at or j ac ob in us 2. 46 4. 27 , 1 .5 6 3. 75 2. 41 , 1 .8 6 1. 53 6. 38 5. 08 3. 98 4. 36 1. 09 U F 38 73 0 STEADMAN: Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida 13 record of Europe and North America. The abundant specimens of Arenicolumba prattae from Thomas Farm represent the New World’s earliest record of Columbidae. The only earlier fossil assigned to the Columbidae is the single humerus of Gerandia calcaria from the early Miocene (Aquitanian; ca. 22 ma) of Saint Gérand-le-Puy, France (Lambrecht 1933, Brodkorb 1971, Olson 1985). Continental fossils of columbids that are younger than Arenicolumba, mainly from the northern hemisphere, seem to represent mod- ern genera (Olson 1985, Becker and Brodkorb 1992). Columbids do have an extensive Quaternary fossil record, especially on tropical oceanic islands where con- siderable anthropogenic extinction of species and even genera is documented (Steadman 1992, 2006, Pregill et al. 1994). The family Columbidae is broadly distributed in tem- perate and tropical regions today. Its greatest generic and species-level diversity is in the Australian-Papuan- Oceanic region (Steadman 2006), where the substantial morphological and ecological diversity of pigeons and doves stands in marked contrast to the more uniform morphology and ecology of New World columbids. In eight genetically-based cladograms of living columbid genera by Pereira et al. (2007), Oena and Turtur (the two living genera most similar to the extinct Arenicolumba) are rather derived although often with low posterior probabilities or bootstrap support. Should the phylogeny hypothesized by Pereira et al. (2007) be valid, it would suggest that Arenicolumba is not a basal columbid. In having its proposed closest living relatives confined to Africa, Arenicolumba resembles the primi- tive crane Balearica (Gruidae), which occurs in Africa today but is common in the North American Neogene (Olson 1985, Feduccia and Voorhies 1992). Of the nine osteological characters regarded by Livezey & Zusi (2006) to be diagnostic or supportive for defining the Columbidae, only seven (their characters number 1307, 1356, 1369, 1417, 1723, 2036, 2119) could potentially be evaluated in Arenicolumba. I found only one of these characters (number 1723) to be valid and diagnostic of columbids; the remaining six characters are either highly generalized (apply to numerous fami- lies of birds and are unlikely to represent homologies) or cannot be interpreted. At the time of its description in 1992, at least 375 specimens of Arenicolumba prattae had been recov- ered. Subsequent field and laboratory work at Thomas Farm has increased this number to more than 700, which make up 75% to 80% of the site’s identifiable bird fos- sils. It is noteworthy that, among these numerous speci- mens, only the ulnare, manus phalanges, and pedal pha- langes are ever complete. The consistent breakage of other skeletal elements supports the thesis of Pratt (1989) that most breakage of bones in doves (and other small vertebrates) within the Thomas Farm sinkhole was due to damage by avian and mammalian predators at or just before the time of deposition. CUCULIFORMES Traditionally (i.e., Mayr & Amadon 1951, Wetmore 1960), the Cuculiformes was considered to comprise two families, the Cuculidae and Musophagidae. Endemic to Africa today, musophagids (turacos and plantain-eaters) are known as fossils from the Oligocene of Africa and the Oligocene and Miocene of Europe; Ballmann 1970, Feduccia 1996, Mayr 2005). The strictly cuculid char- acters of the five skeletal elements available for Thomasococcyx philohippus (tarsometatarsus, cora- coid, scapula, humerus, and mandible) support the hy- pothesis that the Musophagidae is only distantly related to the Cuculidae and does not belong in the Cuculiformes. This agrees with the proposals of Sibley & Ahlquist (1990), Mindell et al. (1997), Sorenson et al. (2003), Mayr & Ericson (2004), and perhaps Cracraft et al. (2004). I should note, however, that the early Eocene Foro panarium (known from a complete, associated skel- eton, Green River Formation, Wyoming) was placed in an extinct family of Cuculiformes with characters sug- gestive of musophagids, cuculids, and the enigmatic Opisthocomus hoazin (Olson 1992). Furthermore, Hughes (2000) classified the Opisthocomidae (consist- ing of the living Opisthocomus hoazin and the middle Miocene fossil Hoazinoides magdalenae from Colom- bia; A. H. Miller 1953) and Musophagidae together to make up the Opisthocomiformes, this order being sister to the Cuculiformes, consisting only of the Cuculidae. Whether the Cuculidae originated in the New World versus Old World, or Northern Hemisphere versus Southern Hemisphere, awaits the discovery of more Pa- leogene fossils. The Tertiary fossil record of cuckoos is poor. Nearly all of the supposed Eocene and Oligocene cuckoos from Europe are either not referable to the Cuculidae or are of questionable identity; Dynamopterus velox from the Eo-Oligocene Quercy fauna of France may be a cuckoo but is based only on a humerus (Olson & Feduccia 1979, Martin & Mengel 1984, Olson 1985, Mayr 2005, Mayr & Mourer-Chauviré 2005). In North America, Eocuculus cherpinae is based on a partial associated skeleton from the late Eocene (early Chadronian LMA; ca. 37-36 Ma) of Colorado (Chandler 1999). Eocuculus was much smaller than Thomasococcyx and had relatively short legs suggest- ing that it was arboreal rather than ground-dwelling. 14 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 48(1) Chandler (1999) classified Eocuculus not only in the Cuculidae but in the subfamily Cuculinae, with its closest affinities to Cuculus (Old World) among living genera. Mayr (2006) reported Eocuculus from the early Oligocene (ca. 30 Ma) of France, excluded it from crown group Cuculidae, and was justifiably tentative about re- ferring it to the Cuculidae. Neococcyx mccorquodalei is based only on the distal end of humerus from the early Oligocene (late Chadronian LMA; ca. 34 Ma; Prothero & Emry 2004) of Saskatchewan (Weigel 1963). I agree with Olson (1985) that Neococcyx probably is a cuckoo, although the limited material precludes any precise hy- pothesis of relationship. Moving to the Neogene, roughly contemporane- ous with Thomasococcyx is Cursoricoccyx geraldinae, based on a humeral end of coracoid, a nearly complete carpometacarpus, and a proximal end of tarsometatar- sus from the early Miocene (Hemingfordian LMA) Martin Canyon local fauna of Colorado (Martin & Mengel 1984). While I agree with Martin and Mengel (1984) that it belongs in the Neomorphinae, C. geraldinae is much larger than T. philohippus (Table 2) and differs further in four characters of the tarsometa- tarsus (the only directly comparable element in the two species). Finally, Olson (1985:111) mentioned the distal end of a tibiotarsus from “a large cuckoo of uncertain affini- ties” from the early Pliocene (late Hemphillian LMA) Lee Creek local fauna, North Carolina. This specimen (USNM 256228) was tentatively referred to the Galli- formes (family, genus, and species indeterminate) by Olson & Rasmussen (2001), who also noted that this fossil may derive from the early to middle Miocene Pungo River Formation (Barstovian LMA) rather than the Hemphillian Yorktown Formation. The interpretation of Thomasococcyx philohippus as a ground-dwelling species is compatible with the os- teological characters that ally it more closely to living New World genera of ground-dwelling cuckoos (Neomorphinae, which I regard as monophyletic and more closely allied to the Crotophaginae than to the Cuculinae) than to any of the living Old World genera of Cuculidae. Following the classification of Payne (2005), which is based on morphology, DNA, and breeding ecol- ogy, my assignment of Thomasococcyx to the Neomorphinae suggests that the split between the New World ground-dwelling cuckoos and other cuculid sub- families already had occurred by the early Miocene. At this time at least two genera of Neomorphinae (Cursoricoccyx, Thomasococcyx) existed in North America. Today, Geococcyx is the only neomorphine cuckoo north of ca. 24°N, the remaining four genera (Tapera, Dromococcyx, Morococcyx, Neomorphus) being strictly neotropical. I note here that Hughes (2000) placed Morococcyx, Geococcyx, and Neomorphus in the Neomorphinae, but Tapera and Dromococcyx in the Cuculinae. I disagree, as does Payne (2005). It is beyond the scope of this paper to assess the novel hypothesis of Mayr & Ericson (2004) that the Mesitornithidae (endemic to Madagascar; traditionally placed in the Gruiformes) has a sister group relationship to the Cuculidae. Of the 80 osteological characters used by Mayr & Ericson (2004) to propose a close relation- ship between the Cuculidae and Mesitornithidae, only nine (their characters numbered 40, 54, 55, 73-78) can be evaluated in the available material of Thomasococcyx philohippus. In each case, the character state of T. philohippus agrees with that given for modern skel- etons of both Cuculidae and Mesitornithidae by Mayr & Ericson (2004), except that Thomasococcyx agrees with Cuculidae to the exclusion of Mesitornithidae in charac- ter 78. Of the seven osteological characters that Livezey & Zusi (2006) found to be diagnostic or supportive for the Cuculidae (their characters number 1007, 1336, 1572, 1614, 1651, 1658, 1866), none can be evaluated in Thomasococcyx. Nevertheless, I found six of these seven characters to be generalized (applicable to nu- merous families of birds; unlikely to represent homolo- gies) and the other one (1651) impossible to interpret. ACKNOWLEDGEMENTS This research was supported by the National Science Foundation (Assembling the Tree of Life grant DEB- 0228682) and the Toomey Foundation. I thank Richard Hulbert, Bruce MacFadden, Arthur Poyer, and Erika Simons for working with me in the field, generously shar- ing their extensive knowledge of the Thomas Farm site, and facilitating access to the Thomas Farm bird fossils. I also thank the many volunteers who have worked so hard at Thomas Farm through the years, with special recognition to Micah Adams, Norma Jean Aker, Grant Boardman, David Cale, Tabitha Cale, Wiley Dykes, Carolyn Eastwood, Patricia Geisler, Karen Goodheart, Bill Lee, Judy Lundquist, Mary Lynch, Kari Stopp, Bar- bara Toomey, George Weems, and Marcia Wright. The photographs were skillfully taken by Sean Roberts. For comments that improved the manuscript, I thank Rich- ard Franz, Richard Hulbert, Bruce MacFadden, and Storrs Olson. I am grateful to Andrew Kratter, Mark Robbins, Jeff Sailer, Terry Taylor, Tom Webber, and the late Pierce Brodkorb for preparing high-quality modern skeletons of columbids and cuckoos. For access to specimens under their care, I thank Larry Martin (KUVP), Town STEADMAN: Doves (Columbidae) and Cuckoos (Cuculidae) from the Early Miocene of Florida 15 Peterson (KU), and Fred Sheldon (LSUMNS). LITERATURE CITED Ballmann, P. 1970. Ein neuer Vertreter der Musophagidae (Aves) aus dem Chattien von Gaimersheim bei Ingolstadt (Bayern). Mitteilungen der Bayerischen Staatssammlung für Paläontologie und Historische Geologie, 10:271-275. Baumel, J. J., A. S. King, A. M. Lukas, J. E. Breazile, & H.E. Evans, eds. 1993. Nomina Anatomica Avium. Academic Press, London. Becker, J. J. 1987. Neogene Avian Localities of North America. Smithsonian Institution Press, Washington, D.C. Becker, J. J. & P. Brodkorb. 1992. An early Miocene ground-dove (Aves: Columbidae) from Florida. Natu- ral History Museum of Los Angeles County, Science Series, 36:189-193. Brodkorb, P. 1954. A chachalaca from the Miocene of Florida. Wilson Bulletin, 66:180-183. Brodkorb, P. 1956. Two new birds from the Miocene of Florida. Condor, 58:367-370. Brodkorb, P. 1971. Catalogue of fossil birds: Part 4 (Columbiformes through Piciformes). Bulletin of the Florida State Museum, 15:163-266. Brown, R. W. 1956. Composition of Scientific Words. Smithsonian Institution Press, Washington, D.C. Chandler, R. M. 1999. Fossil birds from Florissant, Colo- rado: with a description of a new genus and species of cuckoo. National Park Service, Natural Resources Geologic Resources Division, Geologic Resource Di- vision Technical Report, 99/03:49-53. Cracraft, J., F. K. Barker, M. Braun, J. Harshman, G. J. Dyke, J. Feinstein, S. Stanley, A. Cibois, P. Schikler, P. Beresford, J. García-Moreno, M. D. Sorenson, T. Yuri, & D. P. Mindell. 2004. Phylogenetic relation- ships among modern birds (Neornithes). pp. 468-489 in J. Cracraft and M. J. Donoghue, eds. Assembling the Tree of Life. Oxford University Press, Oxford, U.K. Feduccia, A. 1996. The Origin and Evolution of Birds. Yale University Press, New Haven, CT. Feduccia, A. & M. R. Voorhies. 1992. Crowned cranes (Gruidae: Balearica) in the Miocene of Nebraska. Los Angeles County Museum of Natural History, Sci- ence Series, 36:239-248. Gibbs, D., E. Barnes, & J. Cox. 2001. Pigeons and Doves: A Guide to the Pigeons and Doves of the World. Yale University Press, New Haven, CT. Howard, H. 1929. The avifauna of Emeryville Shellmound. University of California Publications in Zoology, 32:301-394. Hughes, J. M. 2000. Monophyly and phylogeny of cuck- oos (Aves, Cuculidae) inferred from osteological char- acters. Zoological Journal of the Linnean Society, 130:263-307. Hulbert, R. C., Jr. 1984. Paleoecology and population dynamics of the early Miocene (Hemingfordian) horse Parahippus leonensis from the Thomas Farm Site, Florida. Journal of Vertebrate Paleontology, 4:547-558. Hulbert, R. C., Jr., ed. 2001. The Fossil Vertebrates of Florida. University Press of Florida, Gainesville. James, H. F. 2005. Paleogene fossils and the radiation of modern birds. Auk, 122:1049-1054. Lambrecht, K. 1933. Handbuch de Palaeoornithologie. Gebrüder Borntraeger, Berlin. Livezey, B. C. & R. L. Zusi. 2006. Higher-order phy- logeny of modern birds (Theropoda, Aves: Neornithes) based on comparative anatomy: I. methods and char- acters. Bulletin of the Carnegie Museum of Natural History, no. 37. MacFadden, B. J. 1992. Fossil Horses: Systematics, Paleobiology, and Evolution of the Family Equidae. Cambridge University Press, New York. MacFadden, B. J. 2001. Three-toed browsing horse Anchitherium clarencei from the Early Miocene (Hemingfordian) Thomas Farm, Florida. Bulletin of the Florida Museum of Natural History, 43:79-109. Martin, L. D. & R. M. Mengel. 1984. A new cuckoo and a chachalaca from the early Miocene of Colo- rado. Carnegie Museum of Natural History Special Publication, no. 9:171-177. Mayr, E. & D. Amadon. 1951. A classification of recent birds. American Museum Novitates, no. 1496. Mayr, G. 2005. The Paleogene fossil record of birds in Europe. Biological Review, 80:515-542. Mayr, G. 2006. A specimen of Eocuculus Chandler, 1999 (Aves, ?Cuculidae) from the early Oligocene of France. Geobios, 39:865-872. Mayr, G. & P. G. P. Ericson. 2004. Evidence for a sister group relationship between the Madagascan mesites (Mesitornithidae) and the cuckoos (Cuculidae). Senckenbergiana biologica, 84:1-17. Mayr, G. & C. Mourer-Chauviré. 2005. A specimen of Parvicuculus Harrison & Walker 1977 (Aves: Parvicuculidae) from the early Eocene of France. Bulletin of the British Ornithologists’ Club, 125:299- 304. Miller, A. H. 1953. A fossil hoatzin from the Miocene of Colombia. Auk, 70:484-489. Mindell, D. P., M. D. Sorenson, C. J. Huddleston, H. C. Miranda, Jr., A. Knight, S. J. Swachuk, & T. Yuri. 1997. Phylogenetic relationships among and within select avian orders based on mitochondrial DNA. pp. 16 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 48(1) 213-247 in D. P. Mindell, ed. Avian Molecular Evolu- tion and Systematics. Academic Press, San Diego, California. Mourer-Chauviré, C. 1992. Un ganga primitif (Aves: Columbiformes, Pteroclidae) de très grande taille de Paléogène des Phosphorites du Quercy (France). Comptes rendus de l’Académie des Sciences Série II, 214:229-235. Mourer-Chauviré, C. 1993. Les gangas (Aves: Columbiformes, Pteroclidae) de très grande taille de Paléogène et du Miocène inférieur de France. Palaeovertebrata, 22:73-98. Olson, S. L. 1985. The fossil record of birds. Avian Biol- ogy, 8:79-252. Olson, S. L. 1989. Aspects of global avifaunal dynamics during the Cenozoic. Acta XIXth Congressus Internationalis Ornithologici, 2:2023-2029. Olson, S. L. 1992. A new family of primitive landbirds from the Lower Eocene Green River Formation of Wyoming. Natural History Museum of Los Angeles County, Science Series, no. 36:127-136. Olson, S. L. & J. Farrand. 1974. Rhegminornis restud- ied: a tiny Miocene turkey. Wilson Bulletin, 86:114- 120. Olson, S. L. & A. Feduccia. 1979. An Old World occur- rence of the Eocene avian family Primobucconidae. Proceedings of the Biological Society of Washington, 92:494-497. Olson, S. L. & P. C. Rasmussen. 2001. Miocene and Pliocene birds from the Lee Creek Mine, North Caro- lina. pp. 233-365 in C. E. Ray and D. J. Bohaska, eds. Geology and paleontology of the Lee Creek Mine, North Carolina, II. Smithsonian Contributions in Paleobiology, no. 90. Payne, R. B. 2005. The Cuckoos. Oxford University Press, Oxford, U.K. Pereira, S. L., K. P. Johnson, D. H. Clayton, & A. J. Baker. 2007. Mitochondrial and nuclear DNA se- quences support a Cretaceous origin of Columbiformes and a dispersal-driven radiation in the Paleogene. Systematic Biology 56:656-672. Pratt, A. E. 1989. Taphonomy of the microvertebrate fauna from the early Miocene Thomas Farm locality, Florida (U.S.A.). Palaeogeography, Palaeoclimat- ology, Palaeoecology, 76:125-151. Pregill, G. K., D. W. Steadman & D. R. Watters. 1994. Late Quaternary vertebrate faunas of the Lesser Antilles: historical components of Caribbean bioge- ography. Bulletin of the Carnegie Museum of Natu- ral History, no. 30. Prothero & R. J. Emry. 2004. The Chadronian, Orellan, and Whitneyan North American Land Mammal Ages. pp. 156-168 in M. O. Woodburne, ed. Late Creta- ceous and Cenozoic Mammals of North America: Biostratigraphy and Geochronology. Columbia Uni- versity Press, New York. Sibley, C. G. & J. E. Ahlquist. 1990. Phylogeny and Clas- sification of Birds: A Study in Molecular Evolution. Yale University Press, New Haven, Connecticut. Sorenson, M. D., E. O’Neal, J. Garcia-Moreno, & D. P. Mindell. 2003. More taxa, more characters: the hoatzin problem is still unresolved. Molecular Biology and Evolution, 20:1484-1499. Steadman, D. W. 1980. A review of the osteology and paleontology of turkeys (Aves: Meleagridinae). Con- tributions in Science, Natural History Museum of Los Angeles County, 330:131-207. Steadman, D. W. 1992. New species of Gallicolumba and Macropygia (Aves: Columbidae) from archeo- logical sites in Polynesia. Natural History Museum of Los Angeles County, Science Series, 36:329-348. Steadman, D. W. 2006. Extinction and Biogeography of Tropical Pacific Birds. University of Chicago Press, Chicago, IL. Weigel, R. D. 1963. Oligocene birds from Saskatchewan. Quarterly Journal of the Florida Academy of Sciences, 26:257-262. Wetmore, A. 1943. Fossil birds from the Tertiary de- posits in Florida. Proceedings of the New England Zoological Club, 22:59-68. Wetmore, A. 1958. Miscellaneous notes on fossil birds. Smithsonian Miscellaneous Collections, no. 135. Wetmore, A. 1960. A classification for the birds of the world. Smithsonian Miscellanous Collections, no. 139. Winston, J. E. 1999. Describing Species: Practical Taxo- nomic Procedure for Biologists. Columbia University Press, New York.