,f „ T P1 ~U ~L_31 TL :. of the FLORIDA STATE MUSEUM Biological Sciences Volume 26 1980 Number 1 TWO MID-PLEISTOCENE AVIFAUNAS FROM COLEMAN, FLORIDA THOMAS L. RITCHIE r *I S .:Il .... 1&& ; '1 . 1., aa. 0 4 SS S ./ m '. I UNIVERSITY OF FLORIDA GAINESVILLE Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCIENCFS, are published at'irregular intervals. Volumes contain about 300 pages and are not necessarily completed in any one calendar year. OLIVER L. AUSTIN, JR., Editor RHODA J. RYBAK-BRYANT, Managing Editor Consultants for this issue: PIERCE BRODKORB KENNETH E. CAMPBELL STORRS L. OLSON Copyright © 1980 by the Florida State Museum of the University of Florida Communications concerning purchase or exchange of the publications and all manuscripts should be addressed to: Managing Editor, Bulletin; Florida State Museum; University of Florida; Gainesville, Florida 32611. This public document was promulgated at an annual cost of $1573.00 or $1.573 per copy. It makes available to libraries, scholars, and all in- terested persons the results of researches in the natural sciences, em- phasizing the circum-Caribbean region. Publication date: December 22, 1980 Price: 1.60 TWO MID-PLEISTOCENE AVIFAUNAS FROM COLEMAN, FLORIDA THOMAS L. RITCHIEl SYNOPSIS: Avifaunas from two,sites near Coleman, Florida, are Irvingtonian in age and con- tain a total of 38 species. The two sites, C61eman IIA and Coleman IIIC and D, are probably related temporally, but they are analyzed separately. Ten species (Anhinga cf. A. grandia, Buteo lagopus, Aquila chrysaetos, Falco columbarius, Bubo virginianus, Protocitta cf. P. afax, Dendroica sp., Pheucticus ludovicianus, Pooecetes gramineus, and Aimophila aestivalis) are new to the Pleistocene of Florida. Pheucticus ludouicianus and Aimophila aestivalis are new to the fossil record. New material of two populations of Colinus suilium, one early to middle Pleistocene in age (from the Inglis IA site in Citrus County, Florida) and the other middle Pleistocene in age (from Coleman IIA), is analyzed, and indicates C. milium under- went a gradual size increase throughout the Pleistocene. This is the opposite of what was previously thought and suggests that C. suilium was not derived from Colinus hibbardi as has been hypothesized. TABLE OF CONTENTS INTRODUCTION ACKNOWLEDGMENTS. SYSTEMATIC LIST . Coleman IIA Coleman III. C le ° M ~ m ~ 2 * ~ ~ PALEOECOLOGY EVOLUTIONARY TRENDS . MODE OF FOSSIL DEPOSITION . LITERATURE CITED 'The author is currentlv a Naturalist/Guide on board the M.S. Lindblad EXPLORER, as well as a pro- fessional wildlife artist based in Palm Beach, Florida. This study was performed while he was a Graduate Assistant in the Department of Zoology, University of Florida, Gainesville 32611. It represents, in part, a thesis prepared in partial fulfillment of the requirements for the degree Master of Science. RITCHIE, THOMAS L. 1980. Two Mid-Pleistocene Avifaunas from Coleman, Florida. Bull. Florida State Mus., Biol. Sci. 26(1): 1-36. 2 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 INTRODUCTION Two fossil deposits collected near Coleman, Sumter County, Florida, have proved important in the study of middle Pleistocene birds and mam- mals of Florida. The first, Coleman IIA, was discovered in 1966 during limestone mining operations at the Coleman II mine of the Dixie Lime and Stone Corporation. The site was collected by Norm Tessman, S. David Webb, Robert Allen, Charles Collins, and Sue Hirschfeld (Webb 1974) and was destroyed that same year by the ensuing mining opera- tions. The deposit represents a sinkhole filling that extended into late Eo- cene Ocala limestone; it was deposited during the late Illinoian Glacial Stage (Webb 1974). Allen (in Martin i974) gave the following description of the site: The opening of the paleosink was approximately 30 yards long and 25 yards wide. Only the upper 8 to 10 feet of sediments filling the sink were exposed. Thin alter- nating beds of two different types appeared: a wet orange-brown clay and a coarse gray-white sand. Although the clay facies appeared homogeneous, the sand facies were filled with pebble-to-boulder-sized limestone rubble. Bones found in the clay were characteristically colored white or tan, with all grada- tions in between; the bone found in gray-white sand was characteristically black, but in a few instances, white. Bones often transgressed these lithologic units and are characteristically colored half tan and half black, testify- ing to relatively rapid accumulation of successive lithologic units. A study by Tessman (Martin 1974) indicated a large number of the bigger mammalian bones have "green breaks," meaning breaks that oc- curred before fossilization. No evidence indicates carnivores caused these breaks, so it is assumed that fossil remains from some of the larger mam- mals were accumulated by their falling a considerable distance into the sinkhole. To verify this hypothesis he made comparisons of the Coleman IIA fossils with freshly broken bones and bones recently fallen into a sinkhole 65 feet deep. Martin (1974) reported on the mammalian fauna of Coleman IIA, and noted that bat remains, especially Mgotis cf. M. austroriparius, indicated the sink was "extensive and not filled with water." Martin identified 37 mammalian species, including 5 species (Dasypus bellus, Pampatherium septentrionalis, Palaeolama mirifica, Platygonus cumberlandensis, and Arctodus pristinus) indicative of a late Irvingtonian age. 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 3 In 1968 Robert Martin and H. Kelly Brooks discovered and collected another site near Coleman, Coleman III. According to Brooks (pers. comm.) this site was approximately 300 meters due north of Coleman IIA, in the Coleman III mine of the Dixie Lime and Stone Corporation. Unlike the karst deposit of IIA, Coleman III was a small spring deposit that has been dry since the middle Pleistocene (Brooks, pers. comm.). The site was divided into four horizontal sections, A, B, C, and D. A sample from IIIA (the spring mouth) yielded a light gray, fine-grained marl with freshwater snail shell fragments. IIIB (Viviparus zone) consisted of a brownish-gray, fine grained sandstone with slightly darker silt stringers throughout, and many snail shell fragments, mostly of Viviparus georgianus. IIIC is identical to IIIA. Unfortunately no substrate samples were collected from IIID. This site, too, was later destroyed by mining operations. The bird fossils of Coleman III were found only in IIIC and IIID. Fourteen non-avian vertebrates (Amia calva, Micropterus sp., Rana sp., Chelydra sp., Trionyx ferox, Terrapene carolina, Chtysemys concinna, Alligator mississipiensis, Crotalus adamanteus, Pampatherium septen- trionalis, Proboscidea, Hemiauchenia cf. H. macrocephala, Tapirus sp., and Equus sp.) are identified from IIIC. The non-avian vertebrates iden- tified from IIID are Chrysemys scripta, Chr!/semys concinna, Chrysemys cf. C. floridanus, and Sigmodon sp. Both Coleman IIA and Inglis IA, a more recently discovered Pleistocene site of central Florida, have yielded large samples of the ex- tinct quail Colinus suilium. Holman (1961) made a thorough study of the osteology of the New World quails, and the samples of these two popula- tions of C. suilium have added to the knowledge of evolution of Colinus in North America. Inglis IA has been dated to both the late Blancan age (Klein 1971) and the early Irvingtonian age (Webb 1974). No matter which assessment of its age is correct, Inglis IA is considerably older than Coleman IIA. Inglis IA is located along the Gulf Coast near Inglis, Citrus County, Florida, at about the present sea level. Jean Klein and Robert Martin discovered this extremely fossiliferous sandy fissure-fill deposit (Webb 1974), and Klein (1971) made a thorough study of its geology, paleoecology, and ferungulates. Material from the three localities (Coleman IIA, Coleman IIIC and D, and Inglis IA) is presented below. Osteological terminology follows that of Howard (1929). All specimens are in the Florida State Museum. Table 1 lists the bird species from Coleman, Florida. ACKNOWLEDGMENTS I wish to express my deepest thanks to Pierce Brodkorb for his invaluable advice and supervision during this study. H. Kelly Brooks answered numerous questions concerning the 4 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 two Coleman sites and was especially helpful in describing Coleman III. Jon Baskin assisted in the description of the geology of Coleman IIIA, B, and C, and S. David Webb, H. Gregory McDonald, and Michael Frazier provided much helpful aid and technical advice. Financial support was provided through Teaching Assistantships with the Department of Zoology, University of Florida. Essential comparative skeletal material was studied on loan from Pierce Brodkorb of the Department of Zoology, University of Florida, and Storrs L. Olson of the Smithsonian Institution. Table 1. -Pleistocene birds from Coleman, Florida. COLEMAN IIA: Poditymbus podiceps Protocitta cf. P. alax Botaurus lentiginosus Dendroica sp. Ixobrychus sp. Agelaius phoeniceus Ckonia maltha weilld Pandanaris foridana Anas platyrh!/nchos Pheucticus ludoucianus Branta canadensts cf. B. c. hutchinsii Pipilo erythrophthalmus Coragyps occidentalis Passerculus sandwichensis Buteo platypterus Pooecetes gTamineus Buteo jamaicensis Aimophila aestivalis Buteo lagopus Spizella pusilia Aquila chrysaetos Spizella sp. Falco columbarius Emberizinae, genus and species indeterminate Colinus suilium COLEMAN IIIC:Agriocharis anza Rallus elegans Anhinga cf. A. grandis Coturnicops noveboracensis Ardea herodias Philoheta minor Aythya affinis Zenaida macroura Aythya sp. Bubo uirginianus Gallinula chloropus Otus asio COLEMAN IIID:Colaptes auratus Progne subis Agriocharis anza SYSTEMATIC LIST COLEMAN IIA ORDER PODICIPEDIFORMES FAMILY PODICIPEDIDAE Podilymbus podiceps (Linnaeus), Pied-billed Grebe MATERIAL -Tarsometatarsus (UF 11816). REMARKS -Podilymbus is distinguished from Podiceps by the nar- rower and deeper proximal end of the tarsometatarsus; ridges surround- ing both cotylae less pronounced; shaft wider and stouter, with the ridge 1980 RITCHIE: COLEMAN PLEISTOCENE AYIFAUNAS 5 on the internal side of anterior metatarsal groove not as well developed; calcaneal ridges shorter, with inner and outer proximal foramina aligned approximately side by side (in the tarsometatarsus of Podiceps the outer proximal foramen is much lower than the inner one). Table 2.-Tarsometatarsal measurements (in mm), with mean and range, of Recent Podilymbus podiceps podiceps (Brodkorb Collection) and fossil Podilymbus podiceps from Coleman IIA (UF 11816). Proximal Width Proximal Depth RECENT female (5) 7.1 (6.9-7.5) 6.2 (5.8-6.8) male (3) 7.9 (7.8-8.1) 6.8 (6.3-7.3) overall 7.4 (6.9-8.1) 6.4 (5.8-7.3) FOSSIL UF 11816 9.1 7.5 The single tarsometatarsus was not completely ossified before fossilization took place, which shows it was from an immature bird. In spite of this, the specimen is considerably larger than the tarsometatarsi of all recent P. podiceps specimens it was compared with (Table 2). Storer (1976) demonstrated that P. p. magnus, a supposedly large Pleistocene subspecies, is totally untenable, for its size falls within the normal size variation for living P. p. podiceps. The poor state of the f6ssil does not permit comparing its overall size with the measurements Storer (1976) published. ORDER ARDEIFORMES FAMILY ARDEIDAE Botaurus lentiginosus (Rackett), American Bittern MATERIAL.-Tibiotarsus, distal end (UF 11821). REMARKS-The tibiotarsus of Botaurus differs from that of Nycticorax by having shaft stouter and less round in cross section; flange overlap on both internal and external condyles better developed; internal condyle, when viewed medially, more round; anterior intercondylar fossa nar- rower; internal condyle projecting approximately even with external edge of tendinal groove (the condyle is located more externally in that of Nyc- ticorax); depression just proximal to anterior articulating surface of exter- nal condyle less excavated; depression on internal condylar half of anterior intercondylar fossa more excavated; internal ligamental prom- inence more proximally located, generally. 6 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 Ixobrgchus sp. MATERIAL.-Tibiotarsus, distal end (UF 22101). REMARKS.-The tibiotarsus of Ixobrychus is separated from that of Butorides by having flange overlap of condyle poorly developed along outer edges of both condyles; tuberosity on ridge forming internal bound- ary of tendinal groove located closer to supratendinal bridge; flange form- ing posterior ridge of internal condyle less extended, resulting in a nar- rower posterior intercondylar sulcus; ridge separating anterior intercon- dylar fossa from posterior intercondylaf sulcus depressed along external half (this ridge is straighter and more defined in Butorides). The fossil was compared with Recent skeletal material from six species of Ixobrychus: exilis, inuolucris, minutus, sinensis, sturmii, and an- namomeus. It differs from the tibiotarsi of all six species by having the anterior intercondylar fossa more excavated. The fossil is most similar to the tibiotarsi of the first four Ixobrychus species listed above, but is fur- ther separated as follows: In I. exilis the tibiotarsus has less distance be- tween the condyles; condyles relatively thinner; smaller size (Table 3). In Table 3.-Measurements (in mm) of the tibiotarsus of fossil Ixobrychus sp. from Coleman IIA, Recent I. exilis and I. minutus (Brodkorb Collection), and Recent I. in- volucris, I. sinensis, I sturmii, and I. cinnamomeus (Smithsonian Institution). Distal Greatest Depth Greatest Depth Species Width External Condyle Internal Condyle 1. sp 5.6 4.7 f 5,4 I. exilis -- - - mean 5.0 4.5 4.8 range (6) 4.9-5.2 4.2-4.6 4.5-5.0 I. minutus 5.9 5.1 5.3 I. involucris 5.5 4.8 5.1 I. sinensis 5.3 4.5 4.4 I. sturmii 5.9 5.3 5.8 I. cinnamomeus 5.8 5.5 5.8 I. involucris and I. minutus the tibiotarsi have both condyles (anterior view) wider; internal condyles extended in antero-posterior plane (the in- ternal condyle of the fossil extends more medially). In I. sinensis the tibiotarsus has more excavation on external face of external condyle; both condyles (anterior view) wider; relatively greater distance between con- dyles. This fossil tibiotarsus possibly represents a new species, but more material is necessary for an accurate diagnosis. The fossil is most similar to the tibiotarsus of I. exilis in overall characters, and may be simply an aberrant form referable to exilis (Fig. 1). 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 7 C·BAC ·7:3:L 1 1R \ 91//0$'67 ZL1 : 8=f-- .5 I.:f.Ir, -*. I f1:>:'i4:44/SF ;9 1.««--"'-,i~Rneps · , s! 5 mm Figure 1. (A) Anterior and (B) distal views of the tibiotarsus of Ixobryc/ms sp. from Coleman HA (UF 22101). FAMILY CICONIIDAE Ciconia maltha weillsi (Sellards) MATERIAL. -Carpometacarpus, proximal end (UF 11619). REMARKS. -Ciconia is separated from Mycteria by having metacarpal I larger and more robust with shorter process; carpal trochlea wider through lateral extension of external face of trochlea; region just posterior to external ligamental attachment more depressed; anterior carpal fossa located more anteriorly and more deeply excavated; ligamental attach- ment of pisiform process less prominent; flexor attachment longer and more raised from metacarpal II shaft. ORDER ANSERIFORMES FAMILY ANATIDAE Anas platyrhynchos Linnaeus, Mallard MATERIAL.-Tibiotarsus, lacking the proximal end (UF 11642). REMARKS.-Anas rubripes, Anas diazi, and Anas julvigula are here considered subspecies of A. platgrhynchos, following Johnsgard's (1961) suggested classification. Woolfenden (1961) made a thorough study that is useful in separating the various genera within this family, and Ligon (1965) separated A. platgrhynchos from other species of Anas by its larger size. The slimness of the shaft shows it is that of a female. The fossil record 8 BULLETIN FLORIDA STATE MUSEUM Vol. 26,.No. 1 of this species is well documented from Pleistocene and prehistoric sites around the world, including about a dozen sites in Florida alone (Brodkorb 1964) . Branta canadensis cf. B. c. hutchinsii (Richardson), Canada Goose MATERIAL.-Carpometacarpus, proximal end (UF 22103). REMARKS.-The specimen was compared to carpometacarpi of Branta, Anser, Anabernicula, and Brantadorna species, and found to be most similar to those of Branta and Anser. The fossil agrees with Branta in hav- ing a large pollical facet; internal edge of pollical facet extends more mediad; external edge of pollical facet extends more distad (the car- pometacarpus of Anser has a shorter and narrower pollical facet, with the external edge of the facet blunter). The size of the fossil is identical to that of Hutchins' Canada goose, which is among the smallest subspecies of B. canadensis. This subspecies has been recorded from three other Pleistocene sites in Florida: Seminole Field, Melbourne, and Itchtucknee River (Wetmore 1931). ORDER ACCIPITRIFORMES FAMILY VULTURIDAE Coragyps occidentalis (L. Miller) MATE:iuAL.-Coracoid (UF 11799), 2 humeri (UF 11591T, 22102), radius (UF 11593), ulna (UF 22096), 2 tarsometatarsi (UF 11800A, 1180OB), and hallux (UF 22097). REMARKS.-Most of the fossils are fragmentary, so only a few measurements (Tables 4 and 5) could be compared with those published Table 4. -Tarsometatarsal measurements (in mm) of Coragyps occidentalis from seven Pleistocene sites. * indicates mean measurements of Howard (1968). Trochlea Trochlea Trochlea Distal for Digit for Digit for Digit Locality Width II Width III Width IV Width Coleman IIA 16.8 4.8 7.2 4.1 Reddick mean 16.9 4.8 6.7 4.1 range (8) (16.7-17.3) (4.5-5.2) (6.5-6.9) (3.9-4.3) Rancho La Brea 17.5* -- -- -- MeKittrick 18.5* -- -- Carpinteria 17.65* -- -- -- Smith Creek 17.8* -- -- San Josecito 16.8* 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 9 by Howard (1968). Because of the broken nature of the specimens, measurements other than those used by Howard were also made and com- pared with measurements of specimens from the Brodkorb Collection of both C. occidentalis from Reddick, Florida, and Recent Coragyps atratus (Tables 5 and 6). There appear to be no diagnostic characteristics, other than size, (elements of C. occidentalis are larger and/or stouter than those of C. atratus) and intermembral proportions (the wing bones are longer in C. occidentalis, but the leg bones are about the same length in both species) that can be used to separate C. occidentalis from C. atratus. C. occiden- talis was obviously a larger and heavier bird with longer wings and com- paratively shorter legs than C. atratus (Howard 1968). The two species have not been found together in any excavation sites, and it has been suggested that C. occidentalis gave rise to C. atratus (Fisher 1944). Most authorities, including Howard (1968) and Brodkorb (pers. comm.), agree with this hypothesis. It is interesting to note that the two species overlap temporally, but not spatially, in the late Pleistocene and Recent. In western North America, C. occidentalis was displaced, at least in an ecological sense, not by C. atratus, but rather by Cathartes aura (Howard 1968). This apparently did not occur in Florida, because C. atratus and C. aura have been recovered together from several sites of Pleistocene and Recent ages (Brodkorb 1964). FAMILY AccIPITRIDAE Buteo platypterus (Vieillot), Broad-winged Hawk MATERIAL.-Ulna (UF 11823). REMARKS.-The ulna of B. platypterus is separated from that of Buteo lineatus by having internal cotyla rounder; external cotyla smaller; prom- inence for anterior articular ligament larger and more pronounced; olecranon shorter and rounder; impression of brachialis anticus deeper and better developed distally along shaft; external condyle rounder (ex- ternal view) with its edge forming obtuse angle with shaft (the ulna of B. lineatus has the external condyle extending more proximally, forming an acute or right angle with the shaft); ligamental attachment at proximal end of external condyle much more excavated; carpal tuberosity rising much less abruptly from shaft and not projecting as far. Buteo jamaicensis (Gmelin), Red-tailed Hawk MATERIAL.-Tibiotarsus, lacking proximal end (UF 11621), and tar- sometatarsus (UF 11622). REMARKS.-Sexual dimorphism is considerable within this species, so the elements were compared with bones from both B. i. umbrinus and B. i. borealis to try to determine sex. The two fossils are definitely not from 10 B U LLE TIN FLO R ID A S TATE M U SEU M Vol. 26, N o. 1 Table 5.-Measurements of humerus (in mm) of Coragyps occidentalis from seven Pleistocene sites and Recent C. atratus (Brodkorb Collection). * indicates mean measurements of Howard (1968). Width at Width of Shaft Height of Brachial Distal Upper End of above Ectepi- Scar through Midshaft Species Width Brachial Soar Condylar Process Internal Condyle Width C. atratus mean 24.6 13.4 16.9 27.3 10.8 range 23.1-26.3 12.4-15.0 16.3-19.0 25.7-30.3 10.4-11.1 C. occidentalis UF 11591, Coleman IIA 26.6 14.4 18.7 29.6 -- UF 22102, Coleman IIA -- -- -- -- 11.8 933-1601, Tex. Mem. Mus., Friesenhahn Cave, Texas 28.0 14.4 19.0 30.8 -- PB 9003 (juv.), Reddick, Florida. 25.1 14.3 18.2 28.8 --- Rancho La Brea, California. 27.4* -- -- 11.5* MeKittrick, California. -- 11.0* Conkling Cavern, California. 27.5* -_ 11.4* San Josecito Cave, California. 26.4* -- --- 11.5* 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 11 Table 6. -Mea5urements of the ulna, radius, and tarsometatarsus (in mm) comparing Coragyps atratus (Brodkorb Collection) to C. occidentalis from Coleman IIA. These unconventional measurements of the tarsometatarsus were made to facilitate fragmental material. Ulna Midshaft Radius Midshaft Species Width Width Depth C. atratus mean (11) 8.2 5.1 4.9 range 7.2-8.8 4.4-5.9 4.4-5.4 C occidentatis (UF 11593) ---- 5.5 5.9 (UF 22096) 8.2 -- -- Tarsometatarsus Distal End of Ridge Ex- Proximal End of Distal tending from Hypotarsus Metatarsal Facet Species Width Width Depth Width Depth C. atratus mean (11) 16.3 7.8 5.5 8.1 5.1 range (14.9-17.7) (7.1-8.3) (5.6-6.0) (7.5-8.8) (4.6-5.5) C. occidentalis (UF 11800A) 16.8 -- ---- 8.8 5.8 (UF 118008) -- 8.6 5.8 -- -- the same individual, because the tibiotarsus corresponds in size to a small male and the tarsometatarsus is from a much larger bird, most likely a female. Buteo lagopus (Pontoppidan), Rough-legged Hawk MATERIAL.-Humerus, distal end (UF 11823). REMARKS.-The fossil is badly worn, but agrees with the humerus of B. lagopus, and differs from that of B. jamaicensis, by having distal portion of shaft flatter and straighter (ventral view); olecranal fossa more ex- cavated; internal tricipital groove better excavated; external tricipital groove wider. This is the first fossil record of B. lagopus from Florida. Aquita chrysaetos (Linnaeus), Golden Eagle MATERIAL.-Femur, distal end (UF 22095). REMARKS.-The femur of A. chrysaetos is distinguished from that of Haliaeetus leucocephalus by having fibular groove more deeply ex- cavated; fibular condyle extends more laterally and distad, causing distal portion of fibular condyle to be more extensive and pointed; external con- dyle narrower (distal view); depression just anterior to distal-most point of external condyle tends to be more deeply excavated. 12 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 The size of the specimen suggests it is a male. More importantly, this is the earliest record of the genus from the New World, and the first fossil record of A. chrysaetos from Florida. FAMILY FALCONIDAE Falco columbarius Linnaeus, American Merlin MATERIAL.-Humerus, distal end (UF 16704), and ulna, distal half (UF 22099). REMARKS.-Wing elements of F. columbarius are very similar to those of Falco sparverius, but are larger (Tables 7 and 8). Although both species exhibit sexual dimofphism, Recent specimens show no size overlap in wing elements. The fossils are slightly smaller than Recent male F. colum- barius, but are considerably larger than Recent female F. sparverius. The humerus of F. columbarius differs from that of F. sparuerius by having olecranal fossa less excavated near external condyle; attachment site on ectepicondylar prominence larger, more dorsal (external) and proximal in its location on shaft; attachment of pronator brevis larger and more ventral; internal ridge of external tricipital groove less pronounced. The fossil humerus has a narrow shaft like that of F. sparverius, but this character is probably variable in F. columbarius. The ulna agrees with F. columbarius in having trochlea wider; small groove just proximal and anterior to internal condyle more apparent. Table 7.-Measurements of the humerus (in mm) of Recent Falco sparverius sparverius (Brodkorb Collection), fossil F. columbarius from Coleman IIA, and Recent F, columbartus (Brodkorb Collection). Shaft Width just Depth of Depth of Distal Proximal to Ect- External Internal Species Width epicondylar Pro. Condyle Condyle F. sport>erius Male mean 6.7 4.5 4.2 2.0 range (5) 6.4-7.1 4.2-4.7 4.0-4.5 1.9-2.1 Female mean 7.5 4.9 4.3 2.2 range (5) 6.9-7.3 4.5-5.3 4.2-4.3 2.1-2.2 Overall mean 6.9 4.7 4.2 2.1 range (10) 6.4-7.3 4.2-5.3 4.0-4.3 1.9-2.2 F. columbanus Fossil 8.1 5.3 4.8 2.5 Recent Male 8.4 5.4 5.3 2.6 Male 8.6 5.6 5.3 2.7 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 13 It is of interest to note that Ligon (1965) demonstrated that F. apart)enius from the Haile XIB Pleistocene was as large or larger than Re- cent F. apart>erius, while the Coleman IIA F. columbarius specimens are, for the most part, smaller than those of the modern F. columbarius. This is the first fossil record of F. columbarius from Florida. The species was previously known from only two Pleistocene sites in North America: Ran- cho La Brea (L. Miller 1921) and McKittrick (L. Miller 1925), both in California. ORDER GALLIFORMES FAMILY PHASIANIDAE Coli*ms suilium Brodkorb MATERIAL.-Cranium (UF 11604), 2 mandibles (UF 11810, 11811), 3 cervical vertebrae (UF 11604), 2 furculae (UF 11605), 7 scapulae (UF 11610), 13 coracoids (UF 11609), 2 sterna (UF 11606), 18 humeri (UF 11611), 16 ulnae (UF 11814), 4 radii (UF 11613), 9 carpometacarpi (UF 11614, 11813), 5 fused thoracic vertebrae (UF 11615, 11815), 2 synsacra (UF 11608, 11818), 16 femora (UF 11616), 17 tibiotarsi (UF 11617), 9 tar- sometatarsi (UF 11618), 5 ribs, 2 ulnares, 3 proximal phalanges, hallux, and 2 pes terminal digits. REMARKS.-Holman (1961) made a thorough study of the New World quails, including both fossil and living species. He compared 94 Table 8.-Measurements of the ulna (in mm) of Recent Falco sparverius sparverius (Brodkorb Collection), fossil F. columbarius from Coleman IIA, and Recent F. columbarius (Brodkorb Collection). Shaft Width just Depth of Distal Proximal to External Species Width External Condyle Condyle F. spanierius Male mean 4.4 2.9 3.8 range (5) 4.1-4.7 2.8-3.2 3.6-3.8 Female mean 4.6 3.1 3.9 range (5) 4.4-4.7 2.9-3.3 3.8-4.1 Overall mean 4.5 3.0 3.9 range (10) 4.1-4.7 2.8-3.3 3.6-4.1 F. columbanus Fossil 5.1 3.4 4.3 Recent Male 5.3 3.4 4.5 14 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 osteological characters and 15 intermembral proportions for each of the living genera and included as many as possible for the fossil genera. He also listed qualitative characters useful in separating both fossil and living species of Colinus (C. hibbardi, C. suilium, C. uirginianus, C. leucopo- gon, C. cristatus, and C. nigrogularis) and included measurements of all species of Colinus except C. cristatus. Unfortunately all of Holman's specimens of C. suilium were late Pleistocene in age and revealed no physical changes or trends with respect to time. An early Pleistocene site, Inglis IA, contains a very large sample of C. suilium, and along with the specimen from Coleman IIA, has pro- vided important data helpful in tracing a more complete study of the species development in Florida throughout the Pleistocene. In the two populations of C. st£ilium presented here, the major variables (other than the ages of the deposits) are minimized. Both populations are from late in glacial stages, which would tend to decrease any differences that are due to effects of "Bergmann's principle." The two localities where the fossils were collected are near each other, separated by an east-west distance of about 60 kilometers, eliminating any possible north-south size gradient, such as that found today in C. uirginianus (see Holman 1961). The removal of these two factors would make any dif- ferences between the two populations attributable to age, i.e. different stages of evolution. Species of Colinus cannot be separated by intermembral proportions (Holman 1961: 152, 206). Intermembral proportions calculated for the two populations of C. suilium showed no substantial differences, except for a difference of 7 % in the tarsometatarsus/humerus ratios (Table 9). However other leg/wing ratios and wing/leg ratios showed no substantial proportional differences. Therefore the 7 % difference in the tar- sometatarsus/humerus ratios is considered to be insignificant. No reliable osteological characters can be used to separate the two populations, although some slight average differences in intramembral proportions are evident. As mentioned earlier, a gradual size gradient occurs in C. uirginianus from north to south. In the eastern United States, the largest specimens (C. u, virginianus) occur in Michigan and the smallest (C. u. floridanus) occur in southern Florida. Specimens of C. suilium from Arredondo, Williston, Orange Lake, Reddick, and Haile average as large or larger than Recent C. u. uirginianus from Michigan in most measurements (Table 10, and Holman 1961:222-223). Specimens of C. suilium from Coleman IIA and Inglis IA are of interest because they are smaller than those of the later populations of C. suilium, The Inglis quail is about equal in size to C. uirginianus of northern Florida, while the Coleman quail is about midway in size between the quail of northern Florida and C. u. virginianus of Illinois. 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 15 Table 9.-Intermembral proportions (in percent) of Colinus suilium from Coleman IIA and Inglis IA, and overall range for genus Colinus (from Holman 1961: 152). Colinus Coleman IIA Inglis IA Overall wing/leg 65 67 63-69 ulna/humerus 91 88 86-91 tarsornetatarsus tibiotarsus 58 55 57-61 tarsometatarsus femur 79 77 73-83 humerus/femur 84 88 82-88 humerus tibiotarsus 62 63 62-66 ulna/femur 76 78 72-79 udna tibiotarsus 56 56 53-59 ulna tarsometatarsus 97 101 92-104 carpometacarpus fennur 45 46 43-47 carpometacarpus tibiotarsus 33 33 33-35 carpometacarpus tarsometatarsus 57 60 55-62 tarsometatarsus humerus 94 87 86-98 Table 10.-A chronological list comparing mean measurements (in mm) of Colinus suilium from eight Pleistocene sites of central Florida with the number of specimens (in parenthesis) for each. (Arredondo I and II, Orange Lake, and Reddick measurements from Holman 1961; Haile XIB measurements from Ligon 1965; Williston measurements from Holman 1969.) Humerus Locality Length Proximal Width Distal Width Orange Lake 36.40(1) 9.85(4) 7.58(3)a Reddick 35.91(31) 9.75(66) 6.98(84)a Arredondo I 37.75(2) 10.02(6) 7.45(2). Arredondo II -- 10.00(1) 7.10(3)a Haile XIB -- 9.31(15) 6.52(14) Williston 35.84(4) 9.83(4) 7.45(4)b Coleman IIA 34.71(8) 9.35(15) 7.09(14) Inglis IA 34.93(40) 9.07(70) 6.99(68) aHolman (1961) published the distal width as "width below epicondyles". His measurements have been altered to con- form to the greatest distal width published for the other sites, Coleman IIA, and Inglis IA. The alteration involved the addi- tion ~f an average value of 0.35 mm. Holman (1959) did not include the mean value for this measurement, but did give the overall range. From this, a me- dian value was figured and substituted in place of a mean value. 16 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 Table 10 continued Ulna Locality Length Proximal width Orange Lake -- 6.00(1) Reddick 32.33(11) 6.11(23) Arredondo I 33.05(2) 6.30(2) Arredondo II 33.00(1) 5.90(1) Haile XIB 32.42(4) -- Williston 34.00(1) 6.60(1) Coleman IIA 31.56(9) 5.99(11) Inglis IA 30.89(12) 5.72(24) Carpometacarpus Locality Length Proximal height Orange Lake 19.60(1) 5.90(1) Reddick C 19.05(10) 5.63(10) Reddick 18.93(16) 5.75(26) Arredondo I 19.30(1) 5.50(1) Arredondo II - - Haile XIB 18.99(10) 5.87(15) Williston 20.00(1) 5.90(1) Coleman IIA 18.67(5) 5.65(9) Inglis IA 18.22(20) 5.54(22) Scapula Locality Proximal width Orange Lake 7.13(3) Reddick 6.94(7) Arredondo I -- Arredondo II 7.10(1) Haile XIB -- Williston -- Coleman IIA 6.87(7) Inglis IA 6.73(44) Coracoid Locality Length Distal width Orange Lake 29.53(3) 9.00(2) Reddick 28.99(16) 8.79(7) Arredondo I 30.82(2) 9.00(2) Arredondo II 29.80(1) 8.70(1) Haile XIB -- -- Williston -- -- Coleman IIA 28.08(5) 8.75(6) Inglis IA 27.21(49) 8.60(18) 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 17 Table 10 continued Femur Locality Length Proximal width Distal width Orange Lakee 42.40(4) -- 7.07(7) Orange Laked - -- 6.60(1) Reddick 42.41(6) -- 7.00(30) Arredondo I -- -- -- Arredondo II 42.50(1) -- 6.90(1) Haile XIB 7.39(8) 7.01(9) Williston 43.60(1) 7.05(2) Coleman IIA 41.27(7) 6.98(13) 6.84(11) Inglis IA 39.61(12) 7.04(21) 6.68(21) Tibiotarsus Locality Length Proximal width Distal width Orange Lake 58.50(3) 9.93(3) 5.65(4) Reddick A 57.20(1) 9.23(12) 5.58(36) Arredondo I 60.00(1) 9.65(2) 5.45(2) Arredondo II -- -- 5.57(3) Haile XIB -- 9.23(6) 5.33(12) Williston 56.00(1) 8.90(1) 5.50(3) Coleman IIA 56.29(1) 9.27(8) 5.35(7) Inglis IA 55.42(2) 8.83(19) 5.21(21) Tarsometatarsus Locality Length Proximal Width Distal Width Orange Lake 6.15(2) 6.00(3) Reddick 32.87 5.88(24) 6.03(27) Arredondo I 34.60(1) 6.20(2) 6.50(2) Arredondo II -- 6.00(1) -- Haile XIB 33.65(2) 5.80(7) 6.22(12) Williston -- -- -- Coleman IIA 32.48(4) 5.73(9) 6.02(7) Inglis IA 30.46(10) 5.64(9) 5.84(19) ~C)range Lake, Arredondo clay. Orange Lake, Arredondo-Wicomico contact. As Holman pointed out, C. hibbardi, from the upper Pliocene of Kan- sas, is the most distinctive species of the genus because of its large size and qualitative osteological characters (see Wetmore 1944). C. suilium is less distinct from C. virginianus than from C. hibbardi and, as Brodkorb (1960) noted, is probably a temporal equivalent to C. virginianus. The evolutionary trend of species in the genus Colinus has been thought to be simply a gradual size decrease through time, but the new material from 18 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 Coleman IIA and Inglis IA demonstrates this is not correct. It has been hypothesized that C. suilium was derived from C. hibbardi, but consider- ing these new data, it is possible to hypothesize that a different species of Colinus, much smaller than C. hibbardi, gave rise to C. suilium. The im- portant point to be made here is that C. suilium apparently underwent a gradual size increase, rather than a decrease as previously thought, throughout the Pleistocene of Florida. Agriocharis anza Howard MATERIAL.-3 skulls and premaxilla (UF 11588), 58 vertebrae (UF 11597), 18 coracoids (UF 11590), 12 sterna (UF 11589), 37 humeri (UF 11591), 6 scapulae (UF 11595), 23 ulnae (UF 11592), 9 radii (UF 11593), 12 carpometacarpi (UF 11594), 7 synsacra (UF 11598), 27 femora (UF 11599), 33 tibiotarsi and 2 fibulae (UF 11600), 35 tarsometatarsi (UF 11601), 4 manus phalanges, and 16 pes digits. REMARKS. -A study by Steadman (1975) consisted of detailed osteological comparisons of both fossil and living turkeys, and included the specimens of A. anza from Coleman IIA. He made comparisons be- tween Proagricocharis kimballensis, Agriocharis progenes, A. ocellata, A. anza. Parapavo californicus, and Meleagris gallopauo. Steadman con- cluded, on the basis of osteological similarities and because most of the qualitative osteological differences are only average ones, that the genera Proagriocharis, Agriocharis, and Parapauo should be synonymized with the genus Meleagris. Paynter (1955) found Agriocharis not deserving of a separate genus from Meleagris, and Steadman (1975) stated that there is little doubt M. gallopauo evolved from A. anza (or as he referred to it, Meleagris anza). This would certainly indicate that Agriocharis, at least, should be synonymized with Meleagris, but for the purposes of this paper the Coleman turkeys are still referred to Agriocharis. ORDER RALLIFORMES FAMILY RALLIDAE Rallus eleganus Audubon, King Rail MATERIAL.-Humerus, distal end (UF 16703), ulna, proximal end (UF 16702), femur, distal end (UF 11820). REMARKS.-Ligon (1965) listed characters by which elements of the larger rails (Rallus, Fulica, and Gallinula) can be segregated, but he did not include elements of Porphyrula, which also overlaps these rails in size. The humerus of Rallus is separated from that of Porphyrula by hav- ing the distal end narrower; entepicondyle more pointed and located more distally; intercondylar furrow and furrow between entepicondyle 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 19 and internal condyle both wider; internal condyle less constricted ven- trally; ectepicondyle less pointed. The ulna of Rallus differs from that of Porphyrula by having depres- sion just proximal to prominence for anterior articular ligament much more excavated; olecranon more pointed and extending more proximally. The femur of Rallus is separated fom that of Porphyrula by having ridge connecting posterior articular surface of internal condyle to shaft better developed and connecting to shaft laterally; a deep depression on medial side of shaft just posterior to anterior ridge extending from inter- nal condyle (this depression is absent in Porphyrula); rotular groove wider; ligamental attachment just proximal to fibular groove larger. The tibiotarsus of Rallus differs from that of Porphyrula by having ridge connecting anterior face of external condyle to shaft better developed, forming a more gradual anterior extension of condyle; shaft thicker where it meets with internal condyle; shaft more rounded in cross section; external condyle more pointed anteriorly. Coturnicops not)eboracensis (Gmelin), Yellow Rail MATERIAL.-Tibiotarsus, distal end (UF 16709). REMARKS.-Feduccia (1968) compared Coturnicops noveboracensis with C. avita from the Upper Pliocene of Idaho and concluded that C. auita is probably the ancestor of C. noueboracensis. Laterallus exilis oc- curred in Florida during the Pleistocene (Olson 1974), and its elements are similar in both size and characters to those of C. noveboracensis. The tibiotarsus of C. not)eboracensis is separated from that of Laterallus by having intercondylar fossa more narrow; ridges extending posteriorly from both condyles protruding farther from shaft, making posterior intercondylar sulcus more excavated; external and internal ligamental prominences both better developed; bridge over peroneus pro- fundus groove located more proximally; external condyle more rounded (lateral view); a distinct furrow between medial connection of supraten- dinal bridge and medial ridge of tendinal groove (in that of Laterallus the supratendinal bridge connects directly to the medial ridge of the tendinal groove and forms no furrow). C. noveboracensis previously was known as a fossil from the Reddick (Brodkorb 1957) and Haile (Ligon 1965) sites of Florida. ORDER CHARADRIIFORMES FAMILY ~COLOPACIDAE Philohela minor (Gmelin), Woodcock MATERIAL.-Mandible (UF 11796), humerus (UF 11798), 2 coracoids (UF 11797), and tibiotarsus, distal end (UF 16701). 20 BULLETIN FLORIDA STATE MUSEUM Vol. 26, N6. 1 REMARKS.-Ligon (1965) listed osteological characters that separate four genera, all of similar size, of the family Scolopacidae: Philohela, Capella, Limnodromus, and Totanus. This species was previously known from only three Pleistocene sites, including Haile and Eichelberger Cave (Brodkorb 1957) of Florida. ORDER COLUMBIFORMES FAMILY (OLUMBIDAE Zenaida macroura (Linnaeus), Mourning Dove MATERIAL.-Humeri, proximal half and distal end, probably from a single bone (UF 16706), and tibiotarsus, proximal end (UF 22104). REMARKS.-A fossil dove from the Upper Pliocene of Kansas, originally identified as Z. macroura by Wetmore (1944), was later described as a new species, Zenaida prior, by Brodkorb (1969). Brodkorb (1969) separated the two species and stated that Z. prior is allied and apparently ancestral to Z. macroura. ORDER STRIGIFORMES FAMILY STRIGIDAE Bubo virginianus (Gmelin), Great Horned Owl MATERIAL.-Tibiotarsus, external condyle (UF 22098). REMARKS.-The fossil is separated from the tibiotarsus of Strix by hav- ing a well-defined depression just proximal to external condyle. The tibiotarsi of both Bubo and Nyctea have a ridge between the condyles separating the intercondylar fossa into two depressions, with one of them located just proximal to the external condyle (the fossil shows evidence of this). The fossil is separated from the tibiotarsus of Nyctea by having posterior ridge of external condyle wider; a convex medialside and a con- cave lateral side on external condyle (the external condyle of Nyctea has two flat sides converging at the ridge); external condyle less elongated in antero-posterior plane; more Range overlap on lateral face of external condyle. The fossil is more robust than the tibiotarsus of Recent B. uirginianus (Table 11). Husband (1924) demonstrated that the supposed large extinct Bubo sinclairi of late Pleistocene actually falls well within the size range of B. uirginianus of Rancho La Brea, and Howard (1947) suggested B. sinclairi may simply be a geographical variant of B. virginianus From Rancho La Brea. B. sinclairi is probably not a viable taxon, so the Cole- man specimen is referred to B. virginianus. This fossil is particularly important because it represents (1) the first fossil record for B. virginianus in Florida, and (2) the earliest fossil record 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 21 for the genus in the New World. The earliest known species of Bubo is Bubo incertus from the Upper Eocene or Lower Oligocene of France (Brodkorb 1971). Oma asio (Linnaeus), Screech Owl MATEFUAL.-Synsacrum (UF 11608), and tarsometatarsus (UF 11623). REMARKS- 0. asio is similar to Aegoliusfunereus in all elements except for the tarsometatarsus. The tarsometatarsus of Otus differs from that of Aegolius by having a relative length 1/3 longer; shaft much narrower. The neural spines of the fused sacral vertebrae are longer in Otus, giving the anterior dorsal surface a higher, arched appearance. The anterior neural spines are broken away in the fossil, but remnants of the iliac fossae indicate it had the same condition as in Otus. 0. asio has been recorded extensively from the North American Pleistocene (Brodkorb 1971). ORDER PICIFORMES FAMILY PICIDAE Colaptes auratus (Linnaeus), Common Flicker MATERIAL.-Humerus (UF 16708), and carpometacarpus (UF 16707). REMARKS.-All North American flickers are now considered conspecific (Short 1965, Brodkorb 1971). The humerus of Colaptes is separated from that of Melanerpes by having pneumatic fossa deeper, with fewer and Table 11.-Measurements (in mm) of the tibiotarsus of Recent Bubo cirginianus (Brodkorb Collection, fossil Bubo uirginianus from Coleman IIA (UF 22098), and Recent Nyctea scandiaca (Brodkorb Collection). Depth of Depth of Shaft Distal Width of External just above Species External Condyle Condyle External Condyle B. uirginianus Male mean 5.4 11.8 5.0 range (7) 5.0-5.7 11.2-12.3 4.5-5.3 Female mean 6.0 13.1 5.5 range (6) 5.9-6.2 12.6-13.9 5.4-5.7 (UF 22098) 6.6 13.8 6.0 N. scandiaca Male mean 5.65 12.3 4.85 range (2) 5.6,5.7 12.0,12.6 4.7,5.0 Female (1) 5:7 13.6 5.5 22 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 smaller foramina; region about attachment of infraspinatus thicker; deltoid crest larger; impression of brachial anticus larger; entepicondyle more extensive. The carpometacarpus of Colaptes differs from that of Melanerpes by having ligamental attachment of pisiform process larger and more extensive; intermetacarpal tuberosity larger and more distad; process of metacarpal I longer. ORDER PASSERIFORMES Hamon (1964) made a detailed study of osteological characters for those genera of each passerine family recovered from the Pleistocene deposits of eastern North America. The elements he studied are the humerus, ulna, carpometacarpus, coracoid, femur, tibiotarsus, and tar- sometatarsus. Hamon's work proved of great value during the study of passerine birds from Coleman IIA. FAMILY HIRUNDINIDAE Progne subis (Linnaeus), Purple Martin MATERIAL.-Humerus (UF 11625). REMARKS.-The large size of the fossil (length: 2.2 cm) separates it from the humeri of all other North American swallows. P. subis was previously reported from the Pleistocene only from Reddick (Brodkorb 1957). FAMILY CORVIDAE Protocitta cf. P. aiax Brodkorb MATERIAL.-Humerus, distal end (UF 16705). REMARKS.-The specimen is most similar to the humeri of two extinct species of jays, Protocitta dixi and Henocitta brodkorbi, both known only from Pleistocene deposits of Florida. The Coleman IIA humerus differs from that of H. brodkorbi by having the internal condyle more spherical and the entepicondyle located less aconally. The Coleman IIA humerus differs from those of both P. dixi and H. brodkorbi by having ectepicon- dylar prominence more extensive and nearly perpendicular to shaft; ex- ternal condyle wider; internal condyle smaller; external tricipital groove narrower; attachment of anterior articular ligament small; medial spur of ectepicondylar process longer; entepicondyle smaller and thinner; olecranal fossa less excavated into entepicondyle. The projection of the ectepicondylar process and the well developed medial spur are similar to those found in Pica. The Coleman IIA humerus differs from that of Pica by having olecranal fossa less excavated; both tricipital grooves better developed; entepicondyle (ventral view) shorter and wider; attachment of anterior articular ligament smaller. Although 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 23 very little of the shaft remains on the fossil specimen, it was apparently straighter (as in other jays) than that of the humerus of Pica. Brodkorb (1972) described Protocitta ajax from a single tarsometatar- sus and stated that it is stouter than that Of P. dixi. Unfortunately no humeri were recovered. It is quite possible that the Coleman IIA Proto- citta is referable to P. aiax. At this time it is inappropriate to describe a new species on the basis of this single element (Fig. 2). Measurements of the specimen are: distal width 10.8 mm; depth exter- nal condyle 4.6 mm; length external condyle 4.1 mm; and length internal condyle 3.0 mm. P, aiax has been reported from the late Pliocene of both Texas (Palo Duro Falls) and Kansas (Rexroad Ranch) (Brodkorb 1972).lu sys:1.1 , 1 cm Figure 2. (A) Anconal and (B) palmar views of the humerus of Protocitta cf. P. aiax from Coleman IIA (UF 16705). FAMILY PARULIDAE Dendroica sp., Wood Warbler MATERIAL.-Tibiotarsus, distal end (UF 22107). REMARKS.-The tibiotarsus of Dendroica is very similar to that of Geoth4/pis, but differs by having supratendinal bridge wider; external condyle (external view) less round; protuberance located at external foot of bridge less developed; external ligamental prominence less developed. Similarity among the members of this genus prevents specific identifica- tion. This is the first fossil record for Dendroica in Florida. FAMILY ICTERIDAE Agelaius phoeniceus (Linnaeus), Red-winged Blackbird MATERIAL.-Humerus (UF 11808), and tarsometatarsus (UF 11812). REMARKS.-Elements of Agelaius phoeniceus are similar to those of Euphagus carolinus and E. cyanocephalus. The humerus of A. 24 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 phoeniceus is separated from that of E. carolinus by having internal tricipital groove less excavated; brachial depression less excavated; no distinct ridge connecting internal condyle with attachment of anterior ar- ticular ligament: attachment of infraspinatus smaller; region of external condyle and ectepicondylar prominence (distal view) wider; depression on connection between attachment of infraspinatus and internal tubero- sity more developed. The humerus of A. phoeniceus differs from that of E. cyanocephalus by having brachial depression less excavated and shorter; external ridge of brachial depression (ventral view) not straight, but bending where it meets with shaft; depression extending from exter- nal tuberosity onto head less developed. The tarsometatarsus of A. phoeniceus is separated from those of both E. carolinus and E. cyanocephalus by having shaft more robust; metatar- sal facet better excavated; trochlear region (internal view) bending more posteriorly. A. phoeniceus in known extensively from the late Pleistocene. It has been reported from Ontario, California, possibly from New Mexico and Virginia, as well as Seminole Field, Itchtucknee River, Reddick, and Ar- redondo in Florida (Brodkorb 1978). Pandanaris floridana Brodkorb MATERIAL.-Femur, proximal end (UF 22108). REMARKS.-The femur of P. flori(lana is separated from that of Quiscalus mexicanus by having a depression in distal surface of neck underneath head; depression on anterior face just medial to trochanteric ridge less excavated (this depression extends proximally underneath the il- iac facet of the femur of Q. mexicanus); neck generally shorter. The fossil was compared with a large number of specimens of P. floridana from Reddick, now in the Brodkorb Collection. Apparently, P. floridana had a considerable sexual size dimorphism, like that found in Q. mexicanus. The size of the fossil (proximal depth 4.1 mm and proximal width of shaft 3.1 mm) is rather small compared to the Reddick specimens, so it probably represents a female. Brodkorb (1957), in his description of the species, stated that its closest living relatives are prob- ably the two cowbird genera, Molothrus and Tanganavius, but neither of these genera exhibit the size variation found in both P. floridana and Q. mexicanus. P. foridana was previously known only from Reddick and Haile XIB (Ligon 1965). FAMILY F~INGILLIDAE Pheucticus ludovicianus (Linnaeus), Rose-breasted Grosbeak MATERIAL.-Ulna (UF 11806). REMARKS.-The ulna of P. ludovicianus is separable from that of 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 25 Pheucticus melanocephalus by having depression between tricipital at- tachment and olecranon more excavated, with excavation extending into base of olecranon at region of humeroulnar depression narrower and ridged; ridge located ventrally to impression of brachialis anticus gener- ally straight, extending from prominence for anterior articular ligament along shaft (this ridge is more curved in the ulna of P. melanocephalus); distal radial depression more smoothly excavated (it is pit-like in the ulna of P. melanocephalus, with a sharp edge along the distal border). The ulna of P. ludouicianus differs from the ulna of Cardinalis car- dinalis by having entire distal end, including distal end of shaft, com- paratively wider; depression between tricipital attachment and olecranon much less excavated, giving base of olecranon a thicker appearance. The ulna of P. ludouicianus is separated from that of Cardinalis sinuata by having external cotyla much larger; carpal tuberosity smaller. This is the first fossil record for P. ludouicianus. Pipilo erythrophthalmus (Linnaeus), Rufous-sided Towhee MATERIAL.-2 ulnae, distal ends (UF 2210OA and B), and tibiotarsus, distal end (UF 22100C). REMARKS.-Hamon (1964) compared elements from Pipilo with those of Passerculus, Ammodramus, Passerherbulus, and Spizella. For each genus he described 13 given characters for the ulnae and 18 given characters for the tibiotarsi. P. erythrophthalmus has been previously reported from late Pleistocene sites in California and New Mexico, as well as Reddick, Arredondo, and Haile XIB (Brodkorb 1978). Passerculus sandwichensis (Gmelin), Savannah Sparrow MATERIAL.-Ulna, proximal end (UF 11809A), and femur (UF 11809B). REMARKS.-Hamon (1964) compared 13 given characters for the ulnae and 15 given characters for the femora of Passerculus, Ammodramus, Passerherbulus, and Spizella. The ulna of P. sandwichensis is separated from that of Melospiza melodia by having tricipital attachment closer to proximal edge of external cotyla; prominence for anterior articular liga- ment more extensive; external cotyla more extensive at region of proximal radial depression, giving external cotylar edge a more concave shape. The femur of P. sandwichensis differs from that of M. melodia by hav- ing attachment of adductor profundus less developed; attachment of gastrocnemius (external head) smaller and less extensive; internal condyle narrower (less extensive from shaft), making distal end relatively nar- rower; shaft slenderer; attachment of ischio-femoralis smaller; iliac facet narrower anteroposteriorly; trochanter less ridged, giving the proximal end region more rounded appearance; depression located on shaft at base of neck smaller. 26 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 P. sandwichensis has been previously reported as a Pleistocene species only from Reddick (Brodkorb 1978). Pooecetes gramineus (Gmelin), Vesper Sparrow MATERIAL.-Humerus (UF 11804A). REMARKS.-The humerus of P. gramineus is separated from the humerus of Chondestes grammacus by having no well-developed pit be- tween ectepicondylar prominence and shaft; longer and narrower brachial depression; olecranal fossa more excavated, extending entire length of internal condyle (the olecranal fossa only extends from intercon- dylar region to midway along internal condyle in the humerus of C. grammacus); proximal edge of pneumatic fossa I circular; deltoid crest closer to shaft; bone connection between head and internal tuberosity thicker, making capital groove longer. The humerus of P. gramineus differs from that of Passerella iliaca by having distal end relatively narrower; ligamental attachment site just proximad to external condyle less developed; external condyle (external view) less round. The humeri of P. iliaca and C. grammacus are similar in that they are both separated from the humerus of P. gramineus by having medial spurs better developed; pneumatic fossae I and II larger; distal edges of articulating surfaces of heads better developed. Dawson (1948) identified this species from Rancho La Brea; this is its first fossil record for Florida. Aimophila aestivalis (Lichtenstein), Bachman's Sparrow MATERIAL.-Humeri, proximal 3/4 and distal end (probably parts of the same bone) (UF 11803), and 2 carpometacarpi (UF 11801). REMARKS.-The humerus of A. aestivalis differs from that of Melo spiza melodia by having less distance between shaft and ectepicondylar prominence; entepicondyle larger; ectepicondylar prominence smaller; brachial depression shallower; articulating surface of head larger, causing opening of fossa I to be more constricted; internal tuberosity less exten- sive. The humerus of A. aestivalis is separable from that of Junco by hav- ing entire element less robust; proximal and distal ends narrower; shaft lighter; ectepicondylar prominence smaller; brachial depression shallower. The humerus of A. aestiualis differs from that of Aimophila cassinii by having both pneumatic fossae smaller; brachial depression less excavated; external condyle smaller; capital groove longer; entepicondyle shorter; less distance between medial spur and shaft. The humerus of A. aestivalis is separable from that of Aimophila ruficeps by having entire proximal end less robust; attachment of infraspinatus less developed; brachial depression less excavated; external condyle smaller; entepicon- dyle shorter. The humerus of A. aestivalis differs from that of Ammospiza 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 27 maritima by having pneumatic fossae I and II larger and deeper; head larger; entepicondyle smaller; brachial depression larger and deeper. The carpometacarpus of A. aestivalis is separable from that of M. melodia by having intermetacarpal tuberosity much larger; depression at proximal end of shaft of metacarpal II (internal view) more excavated; depression at proximal end of metacarpal III connection with internal condyle more excavated. The carpometacarpus of A, aestivalis differs from that of Junco by having intermetacarpal tuberosity larger; entire element more robust; facet for digit III shorter; flange midway along ten- dinal groove better developed. The carpometacarpus of A, aestivalis is separable from that of A. cassinii by having proximal end less robust; ex- ternal carpal trochlea less extensive; external carpal trochlea (external view) rounder. The carpometacarpus of A. aestivalis differs from that of A. ruficens by having entire element comparatively more robust; metacarpal III more parallel with metacarpal II; facet for digit III shorter. The carpometacarpus of A. aestivalis is separable from that of A. maritima by having intermetacarpal tuberosity larger; facet for digit III more robust; depression between carpal trochlea and metacarpal II (ex- ternal view) more excavated. This is the first fossil record for A. aestivalia. Spizella pusilla (Wilson), Field Sparrow MATERIAL.-Humerus (UF 22109). REMARKS.-Hamon (1964) compared the humeri of Spizella, Passer- culus, Ammodramus, and Passerherbulus. The humerus of S. pusilla dif- fers from that of Spizella passerina by having region between attachment of infraspinatus and internal tuberosity thicker, making pneumatic fossa II smaller; head comparatively smaller; brachial depression shorter. The humerus of S. pusilla is separable from that of Spizella arborea by having entire element less robust; brachial depression smaller and shallower; pneumatic fossae I and II smaller; ectepicondylar prominence less developed. The humerus of S. pusilla differs from that of Spizella pallida by having region between attachment of infraspinatus and internal tuberosity thicker; ectepicondyle smaller; brachial depression shallower. S. pusilla has been recorded as a fossil onlyfrom the Pleistocene of Ar- redondo (Brodkorb 1978). Spizella sp. MATERIAL.-Carpometacarpus, lacking metacarpal III (UF 11805). REMARKS,-The broken nature of the specimen and the great amount of similarity within the various species of Spizella make specific iden- tification impossible. 28 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 Genus and species indeterminate MATERIAL.-Ulna, proximal half (UF 11804B), and ulna, proximal and distal halves to same bone (UF 11807). REMARKS-The specimens are from two different species of sparrows, but because they are both broken more accurate identification is impossi- ble. COLEMAN III ORDER PELECANIFORMES FAMILY ANHINGIDAE Anhinga cf. A. grandis Martin and Mengel MATERIAL-Pit C; ulna, distal half (UF 16664). REMARKE-The ulna of Anhinga is separable from that of Phalacrocorax by having trochlea comparatively longer and narrower; distal articulating surfaces less raised from shaft; radial depression smaller and shallower; internal condylar process and carpal tuberosity less exten- sive; external condylar ridge less round; carpal tuberosity extending nearly parallel to the shaft (in the ulna of Phalacrocorax the carpal tuberosity extends almost perpendicular to the shaft); tuberosity at prox- imal end of internal condyle (internal face) less developed; shaft straighter. In Phalacrocorax, the anconal and inner papilae at each secondary feather region of the ulna are separated by a distance nearly twice that found in Anhinga, which causes the inner papillae to be located on the flat internal surface of the shaft. The specimen differs from the ulna of A. anhinga by size (Table 12) and by having tuberosity on external face of external condyle extending more proximad, makingmost distal papilla closer to tendinal pit; ridge of external condyle attachment to shaft in shape of smooth curve, giving ex- ternal condyle a more elongated shape (in the ulna of A. anhinga the ex- ternal condylar ridge attaches to the shaft at an angle approaching 90 °); articulation surface of external condyle extends more proximad, to most proximal point of ligamental attachment; projection of external condylar ridge (internal face) more anterior and less curved; internal condylar process less developed; carpal tuberosity longer with more anterior pro- jection, forming an acute angle between internal condylar process and carpal tuberosity (this angle is obtuse in the ulna of A. anhinga); ligamen- tal attachment (ventral surface) better developed and more excavated; tuberosity at proximal end (internal face) of internal condyle less developed; distal radial depression better excavated, and its extension more external (Fig. 3). Anhinga grandis, from the late Pliocene of Nebraska, is about the same size as the Anhinga from Coleman, but is known only by the distal 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 29 Table 12.-Measurements of the ulna (in mm) of Recent Anhinga anhinga (Brodkorb Collec- tion) and fossil Anhinga cf. A. grandis (UF 16664) from Coleman IIA. Height of Height of Width Greatest External Internal through Species Distal Width Condyle Condyle Trochlea A. anhinga mean 10.6 7.8 5.5 6.3 range (11) 10.0-11.3 7.6-8.1 5.2-5.8 5.7-6.6 UF 16664 11.8 9.1 6.2 7.6 Length of Carpal Central Length of Tuberosity Ridge from Height of Species Tendinal Pit Intermuscular Line Shaft A. anhinga mean 4.8 6.3 5.1 range (11) 4.3-5.1 5.8-6.5 4.7-5.6 UF 16664 5.9 8.5 5.4 Length from Distal Point of Internal Condyle to Central Distal Part of Carpal Width of Species Tuberosity Shaft A. anhinga mean 3.9 6.2 range (11) 3.5-4.2 5.7-6.6 UF 16664 3.6 6.7 Length from Proximal Edge of Ligamental Attachment Species to Center of Trochlea A. anhinga mean 9.7 range (11) 8.6-10.5 UF 16664 11.1 portion of a single humerus (Martin and Mengel 1975). The distal width of the humerus of A. grandis is 1.2 times as large as that of A. anhinga. This compares favorably to some of the proportions between the Anhinga from Coleman and Recent A. anhinga (Table 13). Rather than describe this fossil as being that of a new species, it is here referred to A. grandis. 30 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 3 cm11 Figure 3. Palmar view of the ulna of Anhinga cf. A. grandis from Coleman IIIC (UF 16664). Table 13.-A comparison of the ratio Coleman IIA Anhinga cf. A. grandis/A. anhinga for the ulnar measurements (from Table 11), showing the inconsistency of the various ratios. Mean Range Distal Width 1.07 1.04-1.18 Height of External Condyle 1.04 1.11-1.20 Height of Internal Condyle 1.13 1.07-1.19 Width of Trochlea 1.21 1.13-1.29 Length of Tendinal Pit 1.23 1.16-1.37 Length of Carpal Tuberosity 1.36 1.31-1.46 Length from Internal Condyle to Carpal Tuberosity 0.92 0.86-1.03 Length from Ligamental Attachment to Trochlea 1.14 1.06-1.29 Width of Shaft 1.08 1.02-1.17 Height of Shaft 1.06 0.96-1.15 ORDER ARDEIFORMES FAMILY ARDEIDAE Ardea herodias Linnaeus, Great Blue Heron MATERIAL.-Pit C; tibiotarsus, distal end (UF 15160). REMARKs.-Ardea herodias ( including what was formerly Ardea oc- cidentalis) is separable from all other North American herons by the much larger size of its elements. A. herodias has previously been reported from Pleistocene sites in California and Oregon,and from Seminole Field, Itch- tucknee River, Melbourne, Bradenton, and Rock Spring in Florida, as well as numerous prehistoric sites (Brodkorb 1963). 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 31 ORDER ANSERIFORMES FAMILY ANATIDAE Aythya affinis (Eyton), Lesser Scaup MATERIAL.-Pit (; humerus (UF 15160). REMARKS--The humerus of A. affinis differs from that of Agthya col- laris by having pectoral attachment larger with more distal extension; in- ternal tuberosity more robust; bicipital crest less curved; angle at which bicipital crest connects to shaft generally obtuse (in A. collaris the angle of thjs connection is approximately 90 °); attachment of anterior articular ligament larger. A. a#inis is usually smaller than A. collaris (Johnsgard 1975), but it tends to have longer wing elements. A. a#inis has been reported exten- sively from Pleistocene sites all over North America (Brodkorb 1964). Aythya sp. MATERIAL.-Pit C; humerus, shaft and partial distal end (UF 22110), carpometacarpus, distal end (UF 22112), and tibiotarsus, distal half (UF 22111). REMARKS.-Ducks show a tremendous amount of intraspecific varia- tion, and many characters are not constant enough to separate closely related species with certainty. The slender shafts of the humerus and tibiotarsus separate these specimens from other genera of ducks. Woolfenden (1961) listed characters by which different genera of ducks are separated. ORDER GALLIFORMES FAMILY PHASIANIDAE Agriocharis anza Howard MATERIAL.-Pit D; femur, proximal half (UF 15151). - REMARKS.-A. anza is the only avian species recovered from both Coleman II and Coleman III. The fossil is identical to the femora of A. anza identified from Coleman IIA. A. anza has been reported from the middle Pleistocene of California and Texas (Brodkorb 1964). It is also known from Inglis IA, but until now that recovery was unpublished. ORDER RALLIFORMES FAMILY RALLIDAE Gallinula chloropus (Linnaeus), Common Gallinule MATERIAL-Pit C; carpometacarpus, lacking metacarpal I and metacarpal III (UF 22113). 32 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 REMARKS.-Olson (1974) synonymized Gallinula brodkorbi with G. chloropus, but noted that G. chloropus in the Pleistocene was larger and heavier than the modern form. This specimen, 35.0 mm in length, falls within the measurements for both Recent and fossil specimens of C. chloropus published by Olson (1974). G. chloropus has been recorded extensively from Pleistocene deposits in Europe, the Middle East, Brazil, Cuba, Idaho, and possibly Arizona, as well as six sites in Florida (Brodkorb 1967). PALEOECOLOGY Martin (1974) stated that Coleman IIA was probably more open and xeric than it is today, and predominantly a savanna. He cited several species of mammals that would indicate this, such as Peromyscus floridanus (sandpine scrub/turkey oak sandhill), Cryptotis parua (pine flatwoods, grass and sedge fields, and Lepus alleni and Platygonus sp. (open and semi-arid = xeric habitats). Many of the birds, especially the sparrows Passerculus sandwichensis, Pooecetes gramineus, Aimophila aestivalis, Spizella pusilia, and Spizella sp., are savanna dwellers, thus bolstering Martin's conclusion. The two most common bird species, Col- inus suilium and Agriocharis anza, if one assumes they preferred the habitats of their living counterparts, give no reason to dispute this. Col- inus uirginianus (C. suilium 9 modern counterpart) inhabits open pinelands and fields, and Meleagris gallopavo (A. anza's modern counter- part) can be found in virtually any type of habitat in Florida (swamps, open woodlands, particularly oak and hickory forest, and clearings). Also, most of the birds of prey (Buteo famaicensis, Buteo lagopus, Aquila chrysaetos, probably Coragyps occidentalis, and Falco columbarius) prefer open country for hunting. The presence of Podilymbus podiceps, Botaurus lentiginosus, Ix- obrychus sp., Ciconia maltha, Anas platgrhynchos, Branta canadensis, Rallus elegans, Coturnicops noveboracensis, and Agelaius phoeniceus suggests a freshwater marsh or pond community must have been nearby. Martin reported finding remains of Sciurus carolinensis and Ochrotomys nuttalli, which indicates that a more mesic habitat was also sampled. Several avian species, most notably Buteo platypterus, Philohela minor, and Pheucticus ludouicianus, support this conclusion. Disregarding extinct species, the avifauna from the Pleistocene of Coleman IIA is very similar to that of central Florida today. Two birds of prey, Buteo lagopus and Aquila chrysaetos, do not normally range as far south as central Florida today, but their overall ranges, especially in eastern North America, are not very different from those of several other 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 33 species of birds recovered from Coleman IIA. Their presence is not suffi- cient to base any significant climatic differences on. One extinct species, Agriocharis anza, has previously been reported only from the Pleistocene of western America. However, this species has a modern counterpart that ranges all across North America, including central Florida, so its presence does not necessarily indicate any affinity to environmental conditions of western America. The paleoecology of Coleman III is more obvious. It was apparently a small, sluggish spring run. All of the avian fossils are of aquatic species, with the exception of a single turkey bone from IIID. Most of the non- avian species (see Introduction for list of non-avian species) are also aquatic, but the presence of Crotalus adamanteus, Hemiauchenia macrocephala, and Equus sp. suggest that a drier, more open sayanna habitat was nearby. Both Coleman IIA and Coleman IIIC and D contained species that in- dicate dry, open savanna habitats were nearby, and both contained numerous aquatic species. The two sites were probably contemporaneous, judging from the extinct species of mammals, Pampatherium septen- trionalis, Equus sp., and possibly Sigmodon sp., and the one extinct species of bird, Agriocharis anza, common to both. If so, the spring of Coleman III was probably the habitat from which many of the aquatic species of birds of Coleman II came. A relatively complete picture of the middle Pleistocene habitat of Coleman can be reconstructed. The area was probably predominantly a savanna or open pineland habitat in which a freshwater spring gave rise to a small sluggish stream that prob- ably had a marsh and river swamp habitat along its run. Farther from the stream was mesic forestland that in turn gave way to the open pineland or savanna. A short distance from the spring, 300 meters south, a sinkhole opened into the underlying limestone of the Ocala group. Whether it formed in the mesic forest or open land is impossible to determine. If this reconstruction is accurate, the Coleman area during the middle Pleistocene was very similar to many of the freshwater spring habitats of central Florida today. EVOLUTIONARY TRENDS It is well documented that Pleistocene representatives of modern species are often characterized by larger size. Gallinula chloropus of Cole- man III is large compared to its modern representative, and Podill~mbus podiceps and Bubo uirginianus of Coleman II are larger than their modern representatives in Florida. Several extinct species, Coragyps oc- cidentalis and Colinus milium of Coleman II, Agriocharis anza of Cole- 34 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 man II and III, and Anhinga cf. A. grandis of Coleman III, are large compared to their modern counterparts. Ixobrychus sp. of Coleman II is larger than a modern species that is closely related or conspecific. Olson (1974) gave other examples of this phenomenon. The one exception to the trend is Falco columbarius of Coleman II, which is smaller than its modern representative of Florida. Two species, Protocitta cf. P. ajax from Coleman II and Anhinga cf. A. grandis from Coleman III, show affinities to the late Pliocene, but neither identification is definite because of a lack of referred material. Analysis of two large samples of Colinus suilium, from Coleman IIA and from the earlier deposit of Inglis IA, has demonstrated that the species underwent an overall increase in size throughout the Pleistocene in Florida. Until now, it was hypothesized that C. suilium was derived from Co~inus hibbardi, a large species from the late Pliocene of Kansas, and gradually decreased in size until the late Pleistocene. C. suilium was prob- ably not derived from C. hibbardi, but from some as yet unknown species, and then increased in size until the late Pleistocene, at which time it most likely gave rise to Colinus virginianus and underwent a size decrease. MODE OF FOSSIL DEPOSITION The mode of fossil deposition for Coleman IIA is unclear. It is possible that roosting or nesting birds of prey were a major source of small animals in this fauna. A significant portion (24 %) of the avifauna in this deposit is comprised of birds of prey. Buteo platypterus and Falco columbarius prey almost exclusively on small birds, Bubo preys commonly on birds of all sizes, and Aquila chrysaetos preys heavily on large birds. Corag!/Ps oc- cidentalis may have scavenged bird remains. It is improbable that birds would have fallen into a large sinkhole and not have been able to escape, unlike many of the large mammals. Food re- mains from predators seem to be a plausible explanation for the concen- tration of avian fossils recovered from Coleman IIA. Bubo virginianus was perhaps most responsible for the extremely varied sample of birds, especially the larger species, because owls commonly roost in caves in Florida and can create tremendous stockpiles of animal bones from their regurgitated pellets. C. occidentalis probably nested on the side of the sink, as some of the recovered fossils of this species are from immature, probably nestling birds, and may also have been a source of animal re- mains. It should be noted that the two most common species from this avi- fauna, Agriocharis anza and Comma milium, are the Pleistocene representatives of two modern species, Meleagris gallopavo and Colinus 1980 RITCHIE: COLEMAN PLEISTOCENE AVIFAUNAS 35 uirginianus, which are known to be prey species of Bubo uirginianus (Bent 1938). No evidence indicates the water birds of Coleman IIA were deposited separately during a time when the sinkhole contained water, because their fossilized bones were recovered in both sand and clay facies along with the remains of other birds and mammals and vary considerably in color. Martin (1974:36) stated the sinkhole was "quite extensive and not filled to the top with water during most of its depositional history" as evidenced by the presence of a relatively large quantity of bat remains. The fossils of Coleman III were not in a single concentration, but spread throughout the area of excavation. Apparently deposition oc- curred quite randomly along the course of the postulated spring run. LITERATURE CITED Bent, A. C. 1938. Life histories of North American birds of prey, Part 2. Bull. U.S. Nat. Mus. no. 170, 482 p. Brodkorb, P. 1957. New passerine birds from the Pleistocene of Reddick, Florida. J. Paleont. 31(1):129-138. . 1960. How many species of birds have existed? Bull. Florida State Mus., Biol. Sci. 5(3):41-53. . 1963. Catalogue of fossil birds. Bull. Florida State Mus., Biol. Sci. 7(4):179-293. . 1964. Catalogue of fossil birds, Part 2 (Anseriformes through Galliformes). Bull. Florida State Mus., Biol. Sci. 8(3): 195-335. . 1967. Catalogue of fossil birds, Part 3 (Ralliformes, Ichthyornithiformes, Charadriiformes). Bull. Florida State Mus., Biol. Sci. 11(3):99-220. . 1969. An ancestral mourning dove from Rexroad, Kansas. Quart. J. Florida Acad. Sci. 31(3):173-176. . 1971. Catalogue of fossil birds, Part 4 (Columbiformes through Piciformes). Bull. Florida State Mus., Biol. Sci. 15(4): 163-266. . 1972. Neogene fossil jays from the Great Plains. Condor 74(3):347-349. . 1978. Catalogue of fossil birds, Part 5 (Passeriformes). Bull. Florida State Mus., Biol. Sci. 23(3):141-228. Dawson, W. R. 1948. Records of fringillids from the Pleistocene of Rancho La Brea. Condor 50(2):57-63. Fedducia, J. A. 1968. The Pliocene rails of North America. Auk. 85:441-453. Fisher, H. I. 1944. The skulls of cathartid vultures. Condor 46:272-296. Hamon, J. H. 1964. Osteology and paleontology of the passerine birds of the Reddick, Florida, Pleistocene. Florida Geol. Surv. 44:1-210. Holman, J. A. 1959. Birds and mammals from the Pleistocene of Williston, Florida. Bull. Florida State Mus., Biol. Sci. 5(1):1-24. . 1961. Osteology of living and fossil New World quails (Aves, Galliformes). Bull. Florida State Mus., Biol. Sci. 6(2): 131-233. Howard, H. 1929. The avifauna of Emeryville Shellmound. Univ. California Publ. in Zool. 32(2):301-394. . 1935. The Rancho La Brea wood ibis. Condor 37(5):251-258. . 1947. A preliminary survey of trends in avian evolution from Pleistocene to Recent times. Condor 49(1):10-13. 36 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 1 . 1968. Limb measurements of the extinct vulture, Coragyps occidentatts, with a description of a new subspecies. Papers Archaeol. Soc. New Mexico 1:115-128. Husband, R. A. 1924. Variability in Bubo cirginianus from Rancho La Brea. Condor 26(5):220-225. Johnsgard, P. A. 1961. Evolutionary relationships among the North American mallards. Auk 78:1-43. 1975. Waterfowl of North America. Indiana Press: Bloomington. 575 p. Klein, J. G. 1971. The ferungulates of the Inglis IA local fauna, early Pleistocene of Florida. M. S. Thesis, University of Florida, Gainesville. Ligon, J. D. 1965. A Pleistocene avifauna from Haile, Florida. Bull. Florida State Mus., Biol, Sci. 10(4):127-158. Martin, L. D., and R. M. Mengel. 1975, Anew species of anhinga (Anhingidae) from the upper Pliocene of Nebraska. Auk 92(1):137-140. . 1925. Avifauna of the McKittrick Pleistocene. Univ. California Publ., Bull, Dept. Geol. Sci. 15(9):307-326. Martin, R. A. 1974. Fossil mammals from the Coleman IIA fauna, Sumter County. In Pleistocene Mammals of Florida. S. D. Webb, Ed. Univ. Presses of Florida, Gainesville. pp. 35-99. Olson, S. L. 1974. The Pleistocene rails of North America. Condor 76: 169-175. Paynter, R. A. 1955. The ornithogeography of the Yucatan Peninsula. Yale Univ., Peabody Mus. Nat. Hist. Bull. 9:1-347. Short, L. L. 1965. Hybridization in the flickers (Colaptes) of North America. Bull. Amer. Mus. Nat. Hist. 129(4)309-428. Steadman, D. W. 1975. The Plio-Pleistocene evolution of turkeys (Aves: Meleagridinae). M.S. Thesis, University of Florida, Gainesville. Storer, R. W. 1976. The Pleistocene pied-billed grebes (Aves: Podicipedidae) Smithsonian Contrib. Paleobiol. 27: 147-153. Webb, S. D. 1974. Pleistocene Mammals of Florida. Univ. Presses of Florida, Gainesville. 270 p. Wetmore, A. 1931. The avifauna of the Pleistocene in Florida. Smithsonian Misc. Coll. 85(2):1-41. . 1944. Remains of birds from the Rexroad fauna of the upper Pliocene of Kansas. Univ. Kansas Sci. Bull. 30, 1(9):89-105. Woolfenden, G. E. 1961. Posteranial osteology of the waterfowl. Bull. Florida State Mus., Biol. Sci. 6(1):1-129. Contributions to the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCI- ENCES SERIES, may be in any field of biology. Manuscripts dealing with natural history of systematic problems involving the southeastern United States or the New Worid tropics are solicited especially. Manuscripts should be of medium length-circa 35 to 200 pages (10,500- 16,000 words). Examination for suitability is made by an Editorial Board. The BULLETIN is distributed worldwide through institutional subscriptions and exchanges. It is considered the responsibility of the author to distribute his paper to all interested individ- uals. To aid in this the author(s) receive(s) 50 copies free, and he may purchase additional separates at cost if ordered when page proof is returned. The author is also responsible for any charges incurred for alterations made by him on galley or page proofs. The Museum will send an invoice to the author for this amount upon completion of publication. PREPARATION OF MANUSCRIPT Contributors should consult recent numbers of the BULLETIN for preferred style and format. Highly recommended as a guide is the CBE Style Manual, 3rd Edition, 1972 (Washington, D.C., Amer. Inst. Biol. Sci.). MSS must be submitted in duplicate (please no onionskin) and satisfy the following minimal re- quirements: They should be typewritten, double-spaced (especially tables, figure captions, and "Literature Cited"), on one side of numbered sheets of standard (8-1/2 x 11 in.) bond paper, with at least one-inch margins all around. Figure legends and tables should be typed on separate sheets. All illustrations are referred to as figures. They must comply with the following stan- dards: Photographs should be sharp, with good contrast, and printed on glossy paper. Drawings should be made with dense black waterproof ink on quality paper or illustration board. All il- lustrations should have a cover sheet. All lettering will be medium weight, sans-serif type (e.g Futura Medium, News Gothic) in cutout, dry transfer, or lettering guide letters. Make allow- ance so that after reduction no lowercase letter will be less than 1 mm high (2 mm is preferred) nor any capital letter greater than 5 mm high. The maximum size for illustrations is 8-5/8 x 14 in. (twice typepage size); illustrations should not be less than typepage width (4-5/16 in.). Designate the top of each illustration and identify on the back with soft pencil by author's name, MS title, and figure number. All manuscripts not submitted in BULLETIN format will be returned to the author for retyping. Manuscripts and all editorial matters should be addressed to: Managing Editor of the BULLETIN Florida State Museum Museum Road University of Florida Gainesville, FL 32611