A New early Pleistocene tapir ( Mammalia : perissodactyla ) from Florida, with a review of Blancan tapirs from the state

MISSING IMAGE

Material Information

Title:
A New early Pleistocene tapir ( Mammalia : perissodactyla ) from Florida, with a review of Blancan tapirs from the state
Physical Description:
p. 67-126 : ill. ; 28 cm.
Language:
English
Creator:
Hulbert, Richard C. Jr.
Publisher:
Florida Museum of Natural History
Place of Publication:
Gainesville, FL
Publication Date:

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
Resource Identifier:
oclc - 698191994
System ID:
AA00000405:00001


This item is only available as the following downloads:


Full Text






FLORIDA
MUSEUM
OF NATURAL HISTORYe


BULLETIN


A NEW EARLY PLEISTOCENE TAPIR (MAMMALIA:
PERISSODACTYLA) FROM FLORIDA, WITH A REVIEW OF
BLANCAN TAPIRS FROM THE STATE


Richard C. Hulbert Jr.


Vol. 49, No. 3, pp. 67-126


2010


UNIVERSITY OF FLORIDA GAl NESVI LLE


UNIVERSITY OF FLORIDA


GAINESVILLE






The FLORIDA MUSEUM OF NATURAL HISTORY is Florida's state museum of natural history, dedicated to
understanding, pres in_- and interpreting biological diversity and cultural heritage.

The BULLETIN OF THE FLORIDA MUSEUM OF NATURAL HISTORY is a peer-reviewed journal that
publishes results of original research in zoology, botany, paleontology, archaeology, and museum science. The
Bulletin is published at irregular intervals, and volumes are not necessarily completed in any one year. Volumes
contain between 150 and 300 pages, sometimes more. The number of papers contained in each volume varies,
depending upon the number of pages in each paper, but four numbers is the current standard. Multi-author issues
of related p'-'."-"-. have been published together, and ;,inj.,.,i;_., about putting together such issues are welcomed.
Address all inquiries to the Managing Editor of the Bulletin.

Jacqueline Miller, Guest Editor for this number
Richard C. Hulbert, Jr., Production Editor for this number

Bulletin Committee
Ann S. Cordell
Richard C. Hulbert, Jr.
Jacqueline Miller
Larry M. Page
Roger W. Portell, h,. ,. I ,1
Irvy R. Quitmyer
David W. Steadman, Lx ,, ,t. i., Member

ISSN: 0071-6154

C,,pi,. i I- 2010 by the Florida Museum of Natural History, University of Florida. All rights reserved. Text,
images and other media are for nonprofit, educational, and personal use of students, scholars and the public. Any
commercial use or republication by printed or electronic media is strictly prohibited without written permission
of the museum.

Publication Date: September 30, 2010







Send communications concerning purchase or exchange
of this publication and manuscript queries to:

Managing Editor of the Bulletin
Florida Museum of Natural History
University of Florida
P.O. Box 117800
(,.',u -.l-, FL 32611-7800 USA

FAX: 352-846-- '`-.7
Email: bulletin@flmnh.ufl.edu









A NEW EARLY PLEISTOCENE TAPIR (MAMMALIA:
PERISSODACTYLA) FROM FLORIDA, WITH A REVIEW OF
BLANCAN TAPIRS FROM THE STATE

Richard C. Hulbert Jr.1

ABSTRACT

Florida has the best late Blancan (early P1l; -r- eevi, 2.6-1.6 Ma) record of 7,',1; i in North America. The
genus is currently known from 25 fossil localities of Blancan age, ranging through most of the peninsular
region of the state. Three species are recognized from this time interval in Florida: the relatively large-sized
Tapirus haysii Leidy; the moderate-sized Tapirus lundeliusi sp. nov.; and a much smaller, as yet unnamed
species. Japirus haysii was previously known from the early Irvingtonian (ca. 1.6 to 1.0 Ma) in Florida;
the specimens reported here from seven localities extend its range into the Blancan, which is in accord
with Blancan records from North Carolina (type locality) and the central United States. The Blancan finds
of T haysii in Florida are all relatively sparse, mostly consisting of one or two isolated teeth per locality.
The unnamed small Blancan species of Japirus is slightly better known: three jaw fragments each with
a single tooth; four isolated teeth; a phalanx; and a navicular from a total of five different localities. It
is significantly smaller than any extant species of Iapirus and all of those previously described from the
Pleistocene of North America. -L' .. ,1" tapirs are known from the Hemphillian of the southeastern United
States (e.g., 7,'q,1 ,, polkensis) and the late Miocene of Europe (e.g., Japirus pannonicus), but this is the
first case of their persistence into the Pleistocene. The small Blancan tapir is on average I: htII;l larger than
7,\'it ,, polkensis, and is the same size as an undescribed .&- er -; of hapirus present in the late Hemphillian
Palmetto Fauna of central Florida.
7,'or; ,, hmdeliusi sp. nov. is described on the basis of multiple skeletons from the Haile 7C and 7G
localities of north-central Florida, and more limited samples from nine other Blancan sites. It apparently
became extinct near the Blancan-Irvingtonian boundary, with its youngest confirmed record at the Inglis
1A locality. Tapirus lundeliusi is of moderate size for the genus, similar to extant Tapirus lerrestris and
7,'it ,r pinchaque, and the middle to late Pleistocene species lapirus veroensis. It is significantly smaller
than Tapirus haysii. The skull of T lundeliusi differs from those of T veroensis and T. haysii in retaining a
number of primitive features, such as having a small maxillary flange, longer nasal bone, shorter lacrimal
bone, and ontogenetically rapid fusion of interparietal to occipital. Derived features of T. lundeliusi include
a reduced frontal shield and increased relative mandibular condyle height. Phylogenetic analysis places
T. lundeliusi as the sister taxon of T. veroensis + T haysii, and the clade formed by these three North
American fossil species is formally designated as Tapirus (Helicotapirus) subgen. nov. Among other
*p.....-.- of Tapirus, the extant T. bairdii and the late Neogene Tj. 1 '-11. are most closely related to T.
(Helicotapirus). They all share a large, triangular interparietal; broad, flat posterior lacrimal process; and
extensive meatal fossa on the dorsal surface of the frontal and nasal bones.

Key Words: Tapiridae; Tapirus; new *p ....--, Blancan; Florida; phylogenetic analysis; biochronology


'Florida Museum of Natural History, University of Florida, Gainesville, FL 32611-7800 USA

Hulbert Jr., R. C. 2010. Anew early Pleistocene tapir (Mammalia: ** .. i from Florida, with a review of Blancan tapirs from the state. Bulletin of the
Florida Museum of Natural History 49(3):67-126.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


TABLE OF CONTENTS

Introduction...... ......................................... .. .......... ... 68
M materials and M ethods............................... ................... 72
Fossil localities and Florida Blancan Biochronology ........ 76
System atic Paleontology............................... ................ 78
TI i,. ; :i subgen nov. ............... .................... 78
Tq / *'. lundeliusi sp. nov ...................... ............... 80
lapirus haysii Leidy 1859 ....................... .............. 81
Tq ini .-- subgenus and species indeterminate ...........86
Description of 7.'q i lundeliusi.................... .............. 87
Description and Comparisons of Tapirus sp. from
Sommers Pit, Walden Pit 1, Kissimmee River 6,
Hubbard Pit, and US 19 Bridge Site...................... 109
Phylogenetic Relationships.......................... ................. 111
D iscu ssion .................................................... . ........... 113
C onclusions...... ......................................... ........ .... ... 116
A cknow ledgem ents...... .... ................. ................. 116
Literature Cited ...... ... ...................... 117
A pp en dix 1 ................................................ .. ........... ... 12 1
A appendix 2 .......................................... ... ............ ... 123
A appendix 3 ................................................ .. ........... ... 126


INTRODUCTION
Fossils referred to the extant genus of tapir,
Tapirus Brisson 1762, are relatively abundant
in Irvingtonian and Rancholabrean faunas in
Florida, and have received considerable study
(e.g., Sellards 1918; Simpson 1929 1945; Bader
1957; Ray 1964; Lundelius & Slaughter 1976; Ray
& Sanders 1984; Hulbert 1995; Graham --. .,).
Two species are currently recognized from Florida
during this interval, Tapirus haysii Leidy in the
early Irvingtonian and Japirus veroensis Sellards
in the late Irvingtonian through Rancholabrean
(Hulbert 1995). The situation is quite different for
the preceding Blancan land mammal age (4.4-1.6
Ma) in Florida, for which there are no formally
described species of ]apirus.
Webb (1974) was the first to indicate the
presence of Tapirus in the Blancan of Florida,
listing records in a table from three localities,
Santa Fe River 1B, Haile 15A, and Inglis 1A
(Fig. 1; Appendix 1), but no descriptions were


provided nor identifications to the *p.....- -. level.
Ray et al. (1981) and MacFadden and Hulbert
(2009) noted that Webb's (1974) list of Santa Fe
River 1B mammals is actually a composite listing
of Blancan mammals (or those thought at the time
to be Blancan) from all Santa Fe River fossil sites,
not just the 1B locality. Furthermore, as detailed
by MacFadden and Hulbert (2009), the Santa Fe
River IB locality actually produced both Blancan
and Rancholabrean vertebrates, contrary to the
assertions of Webb (1974), who was followed by
Kurten and Anderson (1980), that site contained
only Blancan f, V, 1i-
Two size classes of lapirus have been found
in the Santa Fe River 1 region (Fig. 1): a rarer,
larger form that is within the size range of Tp, i,, ,
haysii and a much more numerous group of smaller
teeth and bones that is within the size ranges of both
the Rancholabrean 7apirus veroensis and the new
Blancan species described below. The Santa Fe
River 1 and IB specimens of T. haysii are all isolated
teeth (Fig. 2A-C). Their primary significance is to






HULBERT JR.: Blancan Tapirus from Florida


1,2,23
FLORIDA BLANCAN SITES WITH TAPIRUS 24
1 Haile 7C 4'-= 25
2 Haile 7G
3 Withlacoochee River 1A 5-7, 16
4 Waccasassa River 9A
5 Inglis 1A 3
6 Inglis 1B .-
7 Inglis 1D 14-15
8 Santa Fe River 8
9 Santa Fe River 8A
10 Santa Fe River 8C
11 Walden Pit 2
12 Santa Fe River 1 11, 19,20
13 Santa Fe River 1B
14 Kissimmee River 2 17, 21
15 Kissimmee River 6
16 Inglis 1C
17 De Soto Shell Pit 5 22
18 Devil's Elbow 2
19 Sommers Pit
20 Walden Pit 1
21 De Soto Shell Pit 2
22 Lehigh Acres Pit
23 Haile 15A
24 US 19 Bridge Site
25 Hubbard Pit .


Figure 1. Map of Florida showing location of fossil localities of Blancan age that have produced fossils
of Tapirus. See Appendix 1 for detailed geographic information. Tapirus lundeliusi sp. nov. was found at
localities numbered 1 through 11; Tapirus haysii at localities 12 through 18; and Tapirus sp. indeterminate
at localities 15 and 19 through 25.


show that T haysii was present in the late Blancan
of Florida (contrary to Hulbert 1995). The smaller
teeth of Tapirus from the Santa Fe River 1 region
probably represent a mixture of Rancholabrean
T veroensis and the new Blancan species, but
most isolated teeth of the two species can not be
classified with confidence using morphology and
standard linear measurements. Linear discriminant
analyses between the two taxa for each cheektooth


position have misclassification rates ranging from
13 to over 45 percent, with most falling in the
range of 25 to 30 percent. Discriminant scores for
most Santa Fe River 1 teeth grouped them with T.
veroensis, with only a few specimens classified
as the Blancan species. However, a hypothesis
that they are all T. veroensis can not be falsified,
because of the high misclassification rates. For that
reason they are not included in this study.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)


There are no problems with mixing or
reworking of specimens from the two other Blancan
localities with Tapirus listed by Webb (1974), Haile
15A and Inglis lA. However, the single Haile
15A specimen, UF 17468, is a fragmentary lower
cheektooth. As noted by Robertson (1978), it is
adequate to document the presence of the genus in
the fauna, but is not specifically diagnostic. The Inglis
lA sample of Tapirus consists primarily of isolated
postcranial elements, with a few isolated teeth, two
dorsal skull elements, and three partial maxillae. This
sample was also listed by Webb and Wilkins (1984)
and Graham (2003), but has never been described or
figured until now, nor referred to a species.

A B


Two skeletons of Tapirus, a subadult, UF
121736, and a juvenile, UF 162351, were found
during the first major period of excavations at
the late Blancan Haile 7C locality in 1989-1991
(Appendix 1). These were the first skeletons of
tapirs older than late Pleistocene ever found in
Florida, and also preserved the first known skulls
of Blancan tapirs from the state. The skull of UF
121736 differs significantly in a number of features
from Tapirus veroensis. Following its preparation
in the early 1990s, it was evident that it represented
a new species. However, formal description was
delayed for several reasons including, the lack of a
complete nasal bone in UF 121736 (which provide

C


Je


1 cm


Figure 2. Teeth of Tapirus haysii from the late Blancan of Florida, in occlusal view. A, UF 213920, left
Ml orM2; B, UF 224631, right Ml orM2; C, UF 177842, left p4; D, UF 51250, left P4; E, UF 177843,
right M3; F, UF 177844, left P1. Localities listed in text and Appendix 1.


'N ;






HULBERT JR.: Blancan Tapirus from Florida


-braincase (crushed)


Figure 3. Portions of an articulated juvenile skeleton of Tapirus lundeliusi sp. nov., UF 247102, from
Haile 7G. Other portions of this skeleton, including remainder of forelimbs and both hind limbs removed
prior to photography. Degree of association of the tapir skeletons from Haile 7C and 7G ranged from
nearly fully articulated, such as this individual, to those scattered over an area of several square meters
with little or no direct articulation. Abbreviations: C, cervical vertebra; hum, humerus; L, left (unless
followed by numeral); L (followed by numeral), lumbar vertebra; mand, mandible; max, maxilla; R, right;
scap, scapula; T, thoracic vertebra.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


key characters in this genus), the relative young
ontogenetic ages of the two specimens (which
therefore might not display fully adult character
states for some features), and the low sample size
which greatly limited quantitative comparisons
with other species.
S. D. Emslie led a second excavation at Haile
7C in December 1994-January 1995, with the
primary purpose of finding additional specimens
of fossil birds (see Emslie 1998). The third Haile
7C tapir skeleton (UF 160715), and first adult,
was found by volunteer Reed Toomey early in this
dig. Although missing many of the bones of the
manus and pes, UF 160715 is otherwise extremely
complete and well preserved. Its skull preserved
the nasal bones, unlike that of UF 121736. Several
partial skeletons of tapirs consisting of only
postcranial elements were also found during the
1994-1995 excavation. Morgan and Hulbert (1995)
noted the presence of a new species of tapir from
Haile 7C in a brief description of the site. Three
additional tapir skeletons with skulls, two adults
and a juvenile, were collected during the last major
field season at Haile 7C, in the Fall of 2000.
The quantity of known Blancan tapir
specimens in Florida changed dramatically with the
discovery of the Haile 7G locality in the Spring of
2005. Located less than 100 m east of Haile 7C (Fig.
1), Haile 7G had very similar sediments (laminated
blue-gray clays mixed with limestone boulders and
interspersed with lenses and thin layers of sand-size
fragments of limestone), also fi ,'lil produced
associated to fully articulated skeletons (Fig. 3),
and shared most of the fauna of its neighbor (Table
1). While proximity of vertebrate fossil localities in
the Haile limestone mining region of Florida does
not necessarily mean that the sites will be the same
age, it is the case for Haile 7C and 7G. But the Haile
7G locality was of greater size and produced vastly
more vertebrate skeletons. From May 2005 to April
2008 we collected over 500 tetrapod skeletons at
Haile 7G, and Tapirus was the most common
mammal at the site, represented by over 70 partial
to nearly complete skeletons, all representing the
same .- --.. found at Haile 7C. As of 2010, about
half of the Haile 7G skulls and mandibles had been


prepared and are available for study, and many
of these are juveniles. But the combined sample
from Haile 7C and 7G, augmented by those from
Inglis lA, Santa Fe 8, and elsewhere, is exceeded
among fossil tapirs in terms of overall numbers
of skeletons and large samples of all ontogenetic
stages of skeletal and dental development only by
that of Tapirus I -,A11 u ,, (Hulbert et al. 2009).
The purposes of this study are to formally
name and describe this new ....-p -. of Tapirus,
concentrating on characters of the skull and
dentition, and to place it in a phylogenetic context
relative to Iapirus veroensis, Iapirus haysii,
and other species in the genus. A comprehensive
quantitative study of the Haile 7G sample (including
postcranial skeletal elements) will be done upon
its complete preparation and curation. Ecological
analyses using stable oxygen and carbon isotopes
of the Haile 7G sample of lapirus are on-going
(e.g., De Santis & MacFadden 2008). Other late
Blancan species of 7i'qi in Florida are also
briefly reviewed.

MATERIALS AND METHODS

All measurements are reported in millimeters.
Measurements and general methods are the same
as described in Hulbert (2005) and Hulbert et al.
(2009). To the comparative samples of 7apirus
veroensis used in Hulbert (2005), I have added
specimens accessioned into the UF collection
since 2005, most notably a partial mandible (UF
225215) and isolated teeth (UF 225221-225276)
from Steinhatchee River 2, and partial mandibles
from Jacksonville Beach (UF 223228), Bedman
Creek (UFT 223709), Peace River (UF 227141),
and Dickerson Coquina Pit (UF 244022). The
Hulbert (2005) sample of Tapirus haysii has
been augmented by recently donated specimens
from Leisey Shell Pit 3 (UF 241860, 241870,
241874-241875, 241948-241949, 242001,
243500-243513). The opinion of Ray and Sanders
(1984) that 7' haysii is a valid name and the senior
synonym of Tq i'rus copei Simpson is followed
here (see also Hulbert 1995, 2005; Sanders 2002;
Graham 2'. 9). Dalquest and Schultz (1992) and
Smith and Cifelli (2000) promoted the continued






HULBERT JR.: Blancan Tapirus from Florida


use of T. copei instead of T haysii, but without
citing any evidence, morphologic or otherwise, to
disprove the convincing argument made by Ray
and Sanders (1984).
The recent decision by the International
Commission on Stratigraphy to move the Gelasian
Age to the Pleistocene Epoch (Gibbard et al. 2010),
and thus changing the beginning of the Quaternary
(and Pl;r,- e,,) from about 1.81 to 2.58 Ma, is
followed here. Boundaries and subdivisions of
NorthAmerican land mammal ages generally follow
Bell et al. (2004), with the Blancan/Irvingtonian
boundary at about 1.6 Ma and the Irvingtonian/
Rancholabrean boundary at about 0.3 Ma. Thus
the later portion of the Blancan land mammal age
now falls within the early Pleistocene, not the late
Pliocene. All of the Florida fossil localities under
study here fall in the later portion of the Blancan,
within the interval of about 2.6 to 1.6 Ma. As in
Hulbert (1, 3) and Hulbert et al. (2009), the genus
i7,/'qi; ;s defined as all species ofTapiridae sharing
a closer common ancestry with the type species of
the genus, Japirus terrestris (Linnaeus), than with
the type species of all other valid genera in the
Tapiridae as defined and used by Colbert (2005).
Phylogenetic analyses were done using a
set of 22 cranial, four mandibular, and 13 dental
characters (Appendix 2) for nine ."' .. of New
World 7.'tii'us (Appendix 3). The late Miocene
Chinese species 7Twi i, ... ..;, ..;-'". was
included in one analysis to determine its effect
on the relationships between New World Tapirus.
Character states for T h, -h,. ,i... were taken
from Deng et al. (2008). To overcome problems with
incompleteness, a composite outgroup was made
using states from the four best known early Miocene
tapirids, ,,it, pirus harrisonensis Schlaikjer,
Nexuotapirus marslandensis (Schoch & Prins),
. ,, ii : ) ,ii (Matsumoto), and Paratapirus
helveticus (von Meyer). The character state matrix
is shown in Appendix 3. Most parsimonious trees
were computed by PAST, version 1.91, using the
branch-and-bound algorithm and separate analyses
with ordered and unordered multi-state characters.
INSTITUTIONAL ABBREVIATIONS
UF, Florida Museum of Natural History,


University of Florida, Gainesville, Florida, USA;
UF/TRO, Timberlane Research Organization
collection, now housed at the Florida Museum of
Natural History; USNM, U.S. National Museum,
Smithsonian Institution, Washington, D.C., USA.
MORPHOLOGICAL ABBREVIATIONS
L, greatest length; DL, postcanine diastema
length; HT, height; AW, greatest anterior width
measured across the protoloph or protolophid near
the base of the crown; PW, greatest posterior width
measured across the metaloph or hypolophid; W,
width; i, lower incisor; I, upper incisor; c, lower
canine; C, upper canine; m, lower molar; M, upper
molar; p, lower premolar; P, upper premolar (a
numeral following a tooth abbreviation indicates
a specific tooth locus; e.g., m2 is a second lower
molar). A d or D in front of a tooth abbreviation
indicates a deciduous tooth (e.g., DP1).
STATISTICAL ABBREVIATIONS
x, sample mean; s, sample standard deviation;
MIN, minimum value observed in a sample;
MAX, maximum value observed in a sample; OR,
observed range of a sample; N, sample size; CV,
sample coefficient of variation; p, probability.
ONTOGENETIC DESCRIPTORS
Hulbert et al. (2009) 1, ..iT. ,t J, seven general
ontogenetic life history stages in T.,i based on
tooth eruption and wear. Their temporal durations
are not equal, but their use allows comparisons
between ontogenetically equivalent individuals.
* Very young juvenile: DP1-DP3 and dp2-dp3
with little or no wear are the only fully erupted
cheekteeth; DP4 and dp4 may be erupting.
* Young juvenile: DPI-DP4 and dp2-dp4 are
all fully erupted; MI and ml may be ,..i .i
teeth slightly worn.
* Juvenile: DPI-MI and dp2-ml are all fully
erupted and in wear. Crowns of adult premolars
and second molars fully formed in crypts.
* Subadult: PI-P3, DP4, Ml and p2-p3, dp4,
m I are fully erupted and in wear; M2 and m2
maybe, niptmn. little or no wear on PI-P3 and
p2-p3, heavy wear on DP4 and dp4.
* Young Adult: P4, M2, p4 and m2 have erupted
and are in wear; M3 and m3 either erupting or









Table 1. Mammals present at Florida late Blancan (early Pleistocene) vertebrate fossil localities with Tapirus 7'' ',.''. sp. nov. (excluding
Chiroptera and Cetacea). With = Withlacoochee River; Wacc = Waccasassa River; SFR = Santa Fe River. An 'X' indicates a well confirmed
occurrence at a site while a '?' indicates a possibly occurrence but one whose species-level identification is uncertain either due to
incompleteness of available specimens or absence of needed comparisons with related species. After Morgan and Hulbert (1995), Morgan
and White (1995), Ruez (2001), Hulbert et al. ''-, :), Hulbert and Morgan (2009), and Tedford et al. (2009).


Dasypus bellus
Pachyarmatherium leiseyi
Holmesina '. .. ..
Glyptotherium arizonae
A fegalonyx leptostomnus
Paramylodon harlani
Eremotherium eomigrans
Canis edwiardii
Canis lepophagus
Urocyon citrinus
Procyon rexroadensis
Procyon lotor
Arctodus pristinuis
7 ', macrodon
Satherium piscinarium
Lontra canadensis
Mustelafrenata
Spilogale putorius
Lynx rufus
Lynx rexroadensis
AMiracinonyx inexpectata
Smilodon gracilis
Xenosmilus hodsonae

Scalopus aquaticus
Blarina carolinensis
C. '. parva
Sylvilagus webbi

Lepus sp.


Walden Haile
Pit 2 7C
X


Haile
7G
X


X X X
X X
X X
x

X X


With Wacc
1A 9A
X X

X


X ?


X
X ?


SFR 8 SFR SFR Inglis
8A 8C 1A
X

X
X X X X
X
X X X X
X
X


lB


?
X X
X X X X


? X X

? x x x
? X X X
X 2


Inglis
ID
X










Table 1. Continued.


Walden Haile Haile
Pit 2 7C 7G
X


With Wacc SFR 8 SFR
1A 9A 8A


SFR Inglis Inglis Inglis
8C IA 1B ID


i: ', .. 11. sp .
Sciurus carolinensis
Glaucomys n. sp.
Castor californicus
O .. propinetis
Unidentified Peromyscini
Baiomys sp.
Peromyscus hagermanensis
Peromyscus sarmocophinus
Reithrodontomys wetmorei
Neotoma sp.
Sigmodon medius
Sigmodon curtisi
Ondatra idahoensis
Erethizon poyeri
Erethizon kleini
Neochoerus dichroplax
Tapirus hlundeliusi
Nannippus peninsulatus
Cormohipparion emsliei
Equns sp.
.- I. '. bicalcaratus
MAvlohvus floridanus
Hemiauchenia macrocephala
Hemiauchenia gracilis
Palaeolama mirifica
Odocoileus virginianus
Capromeryx arizonensis
AMammut americanum
Cuivieronius tropics


x x


x
x x


x
x
x x x x
X X X X


x x x x
x


? ?


X X X
x


X X X
x x x
x


x
x x x x


X ?


x x






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


fully erupted, but with little or no wear.
* Full Adult: Lophs of M3 and m3 moderately
worn, but no exposed dentine.
* Old Adult: Lophs of M3 and m3 have exposed
dentine because of heavy wear.
FOSSIL LOCALITIES AND FLORIDA
BLANCAN BIOCHRONOLOGY
The specimens of the new tapir species
were collected from ten different localities in
central Florida and one in southwestern Florida
(Fig. 1; Table 1; Appendix 1). Haile 7C and Haile
7G produced the majority of specimens. Both
are relatively large (> 25 m width; > 5 m depth)
karst solution features filled with layers of clay,
carbonate sand, and boulders that are interpreted
to be remnants of a once much larger and deeper
sinkhole lake (Morgan & Hulbert 1995; Emslie
1998; Hulbert et al. 2006; Morgan & Emslie 2010).
In addition to sedimentology (Shunk et al. 2006),
this hypothesized depositional environment is
supported by the frequent preservation of complete
or nearly complete vertebrate skeletons (a very
rare occurrence for Florida karst fossil sites; Fig. 3)
and very abundant freshwater taxa including fish,
frogs, turtles, alligator, and waterfowl. Tapirs were
the most common large mammal at both Haile 7C
and 7G. Even though Pleistocene sites in Florida
are renown for their relative abundance of tapirs
(e.g., Ray & Sanders 1984), they are typically far
outnumbered by other medium-sized ungulates
such as Equus, Palaeolama, Hemiauchenia,
Odocoileus, and/or Plt h'monus. The extraordinary
abundance of tapirs at Haile 7C and 7G is likely
a combination of a vast area of their preferred
forested habitat (DeSantis & MacFadden 2007) to
generate a large standing population plus an as yet
undetermined taphonomic condition to produce the
concentrations of skeletons (Hulbert et al. 2006).
The fossils of the new species of Tyq i ii,. from
the remaining nine localities (Appendix 1) are less
complete and numerous than at Haile 7C and 7G,
and thus serve primarily to supplement the Haile
samples and provide information on the chronologic
range of the new species. Withlacoochee River 1A
is a previously unreported submerged locality (Fig.


1). I ,-,.---i!- were recovered over a 30 m stretch of
the river channel, both lying loose on the surface
and excavated directly from shallow pockets of
gray sandy clay in the Eocene limestone bedrock.
The mammalian fauna has a high diversity of
xenarthrans, carnivorans, and ungulates (Table 1).
The single tapir tooth recovered from this locality,
a partial M3, is slightly larger than any found at
Haile 7C or 7G. But it is much smaller than ?.." F, of
T. haysii, and within the size range of T. veroensis.
Because the latter contains a much larger sample
size than currently known for the new Blancan
species, this suggests that the potential size range of
the new species should include specimens as large
as the Withlacoochee River 1A tooth. Waccasassa
River 9A (Fig. 1; Table 1) is another previously
unpublished Blancan local fauna. Vertebrate fossils
were collected from a river bank in association
with a molluscan fauna equivalent to that of the
Caloosahatchee Formation in southwest Florida.
Fossils of tapirs were collected both from the
riverbed in the general Santa Fe River 8 area and
from two of the five known in situ deposits in the
area, the ones designated Santa Fe River 8A and 8C
(Fig. 1; Table 1). In contrast with the fossiliferous
deposits from the much more widely studied
Santa Fe River 1 region (Morgan & Hulbert 1995;
MacFadden & Hulbert 2009), those from Santa
Fe River 8 are apparently in situ early Pleistocene
sites, and not late Pleistocene or Holocene sites
containing reworked Blancan fossils. The only prior
published description of a Santa Fe 8 specimen was
in Ray et al. (1981).
Fifteen separate fossiliferous karst deposits
about 2 km southwest of the town of Inglis (Fig. 1;
Appendix 1) were exposed during excavations for
the Cross Florida Barge Canal. Of these, only four
produced fossils of Ti ,ui n., T. haysii at Inglis 1C
(Ruez 2001), and the new species at Inglis lA, 1B,
and ID. Inglis lA has the largest and most diverse
fauna of the Inglis sites (Table 1), and has the best
sample of the new tapir excluding the two Haile
sites. Members of its fauna have been the subjects
of numerous publications (e.g., Meylan 1982;
Morgan & Hulbert 1995; Emslie 1998; Tedford et
al. 2009; Morgan & Emslie 2010). Inglis 1B and ID






HULBERT JR.: Blancan Tapirus from Florida


are two of the lesser know Inglis sites. Their faunas
are generally a subset to that of Inglis lA, with a
few exceptions, such as the presence of Mylohyus
at Inglis 1B and Palaeolama at Inglis ID.
Macasphalt Shell Pit was an extensive
commercial sand and shell quarry in Sarasota
County that produced well studied invertebrate
and vertebrate fossil faunas (e.g., Petuch 1982;
Hulbert 1988; Jones et al. 1991; Emslie 1992). The
late Blancan land vertebrate assemblage associated
with the Pinecrest mollusks from this mine is most
similar to the Haile 15A and Santa Fe River 1 local
faunas from northern Florida ( i',-gan & Hulbert
1995). No fossils of tapirs were recovered at
Macasphalt Shell Pit. However, Pinecrest mollusks
were also mined during the 1970s from two nearby,
much smaller quarries, Sommers Pit and Walden
Pit (Fig. 1; Appendix 1). Vertebrate taxa from
Sommers Pit diagnostic of the late Blancan include
i.. ,. m) -'. platymarginata (Weaver & Robertson),
the very small morph of Holmesina floridanus
(Robertson), Neochoerus dichroplax Ahearn
and Lance, Nannippus peninsulatus (Cope), and
Cormohipparion emsliei Hulbert. Two separate
quarries were mined at Walden Pit, here designated
as the more western Pit 1 and more eastern Pit
2, separated by less than 100 m. They are about
1 km north of Sommers Pit. Fossils of terrestrial
mammals were uncommon at both quarries, but
Walden Pit 1 produced specimens of N. dichroplax
and N. peninsulatus, while the very small morph
of H. floridanus, C. emsliei, and N. peninsulatus
were recovered from Pit 2. Specimens of T.
, It'h. ..i' :i i..i were common at both pits. Walden
Pits 1 and 2 and Sommers Pit each produced only a
single specimen of Tapirus, comprising an isolated
lower premolar and two mandibular fragments with
one tooth each. Despite the very limited nature of
these samples, they are important for two reasons.
One of the jaw fragments is the only referred
specimen of the new species from the southern half
ofFlorida. The other two specimens are significantly
smaller than most other known Blancan Tapirus,
most likely a holdover of a Hemphillian species,
and thus bring the total number of late Blancan
t.....--.. of Tapirus in Florida to three. This small


taxon is also present at Kissimmee River 6, where
it co-occurs with Tw K a,. haysii, Hubbard Pit, and
the US 19 Bridge Site (Appendix 1). The latter
two are also previously unpublished late Blancan
localities.
Morgan and Hulbert (1995) proposed that
late Blancan, Irvingtonian, and Rancholabrean
vertebrate faunas in Florida could be grouped into
eight biochronologic assemblages. These are still
valid, although new discoveries have changed
the first or last occurrences of some species.
Because of a revised definition for the base of the
Irvingtonian by Bell et al. (2004), the "earliest
Irvingtonian" assemblage of Morgan and Hulbert
(1995:85) is now the latest interval of the Blancan.
Morgan and Hulbert's (1995) second assemblage
was based solely on the Haile 7C local fauna and
was admittedly poor defined at that time. The
subsequent recovery of the richer Haile 7G and
Withlacoochee River lA local faunas (Table 1)
allows for a much better biochronologic separation
of Florida late Blancan vertebrate faunas into three
drnr assemblages, with approximate numeric
ages of 2.2-2.6 Ma for the early late Blancan, 1.9-
2.2 Ma for the middle late Blancan, and 1.6-1.9 Ma
for the latest Blancan (Hulbert & Morgan 2009).
The early lateBlancan interval is characterized
by the presence of Paramylodon garbanii
(Montellano & Carranza-Castafieda) (proper
name for Florida specimens formerly referred
to the South American species CL ..*,,th. ,;im.
chapadmalense Kraglievich; see Morgan 2008 for
taxonomy), frequent occurrence of the hipparionine
equids Nannippus and Ca i;', I, ,~, ,' and
the very small form of Holmesina floridanus (as
discussed by Hulbert & Morgan 1993). Tapirus is
typically rare or absent at early late Blancan sites
in Florida. Records of Tapirus haysii are limited to
small samples of isolated teeth at Santa Fe River
1 sites, Kissimmee River 2, and Devil's Elbow,
and a metacarpal 5 from Kissimmee River 6
(Fig. 2; Appendix 1). Fossils from Sommers Pit,
Walden Pits 1 and 2, and the US 19 Bridge Site
also fall within this time interval. All records of the
small, unnamed species of 7'q,; ,. belong to this
subinterval of the late Blancan; no younger records






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. t,, )


are currently known. Haile 15A and Lehigh Acres
Pit each produced a single partial tooth diagnostic
only to the family level, but they presumably
represent Tapirus as it is the only known genus
in North America from the late Miocene to the
Pleistocene.
The middle late Blancan interval in Florida
is defined by the first appearance of Paramylodon
harlani (Owen), which replaced the much smaller
P. garbanii of the preceding interval. Other species
that make their first appearances in Florida during
the middle late Blancan are Canis edwardii Gazin,
Sciurus carolinensis Gmelin, !',. i, ,,,..
wetmorei Hibbard, Ondatra idahoensis Wilson,
Erethizon kleini Frasier, and Erethizon poyeri
Hulbert. The last occurrence of '7,.. ,..I, ,i
medius Gidley occurs during or at the end of this
interval. Specimens of Hohnesina floridanus are
significantly larger than those of the preceding
assemblage (Hulbert & Morgan 1993). Haile 7C,
Haile 7G, and Withlacoochee River 1A are the
primary examples of middle late Blancan faunas.
The Waccasassa River 9A and Santa Fe River 8
localities also likely belong in this interval, but they
lack the necessary taxa to differentiate between this
and the succeeding latest Blancan assemblage. For
tapirs, only 7apirus .i"i,,, hr .. is known from the
middle late Blancan. No records of Tapirus haysii
are known from Florida during this interval, but
geographic coverage of known faunas is limited,
so it may have been living elsewhere in the state, or
it could have been temporarily extirpated.
The latest Blancan Florida vertebrate interval
is defined by the first appearance ofSigmodon curtisi
Gidley and Sigmodon minor Gidley (replacing S.
.,...iJ,- of the preceding interval). Other species
that make their first appearances in Florida during
the latest Blancan are Equir, leidyi Hay, / '/7. ';:.. ..
putorius (Linnaeus), Peromyscus sarmocophinus
Ruez, and Ai. 'j,'ti,, texensis Patton. Inglis lA,
Inglis IC, and De Soto Shell Pit are the primary
S..m.i,'.--. of latest Blancan faunas. Both Tapirus
lundeliusi and 17., '-rus haysii are known from this
interval. While T haysii continued into the early
Irvingtonian in Florida (Hulbert 1995), there are no
younger records of T. hi., i..o


SYSTEMATIC PALEONTOLOGY
Order PERISSODACTYLA Owen 1848
Family TAPIRIDAE Gray 1821
Genus TAPIRUS Brisson 1762
Subgenus HELICOTAPIRUS subgen. nov.
Type Species.-Tapirus veroensis Sellards
1918.
Included Species.-Tapirus haysii Leidy
1859; ,Tivu wo ,In;J. hih. isp. nov.
Etymology.-From the Greek helix (spiral,
coil, or curl) plus Tapirus. In reference to the spiral
shape of the prominent meatal fossa on the frontal
and nasal bones of members of this subgenus.
Definition.-All species of Tapirus that share
a closer common ancestor with Tiq ,it -.. veroensis
than they do with Tapirus (Tapirus) terrestris
(Linnaeus 1758), Tqit "... (Pinchacus) pinchaque
(Roulin 1829), Tapirus (/',1,,. !';,, bairdii (Gill
1865), or Tapirus i( .. -Ji!/ i:iii. Desmarest
1819.
Diagnosis.-Species of Tapirus that differ
from lapirus (lapirus) and iA', ,' (I/; i,)
by having anterior process of maxilla visible in
lateral view dorsal to premaxilla; large, triangular
interparietal; extensive meatal fossa on frontal
and nasal whose medial margin lies very close
to midline of skull; lacrimal with weak or absent
anterior process and broad, flattened p,--.t, ri-,
process; relatively shorter postcanine diastema;
relatively greater depth of mandibular ramus
below molars; anterior margin of ascending ramus
projects anteriorly in lateral view; and increased
relative length of p2. Additionally they differ
from Tq ir. (Tapirus) by having a much shorter
sagittal crest. Additionally they differ from hapirus
(Pinchacus) by having shorter nasal and broader
P1, and by lacking posterolabial cingula on upper
molars. Species of 7apirus that differ from lapirus
(T~,;,i!i) by having longer posterior process of
premaxilla; dorsal maxillary flange small or absent,
not extensive to embrace ossified mesethmoid
cartilage; dorsal surface of nasal and frontal on
same plane; long descending nasal process; narrow
-.il.il crest in adults; posterodorsal process of
maxilla widely exposed dorsal to orbit, forming
base of trough for meatal diverticulum; relatively






HULBERT JR.: Blancan Tapirus from Florida


shorter postcanine diastema; relatively greater
depth of mandibular ramus below molars; anterior
margin of ascending ramus projects anteriorly
in lateral view; and increased relative length of
p2. Species of Japirus that differ from 7,' n,.
(Acrocodia) and Tapirus (I .;. ..wf ii i.) by having
narrow sagittal crest in adults; dorsal surface of
nasal and frontal on same plane; greater exposure
of lacrimal in lateral view with at most one visible,
large foramen; and premaxillary-maxillary suture
extends anterior to alveolar margin of C in lateral
view; and by lacking posterolabial cingula on
upper molars. Additionally they differ from I.
(Acrocodia) by having shallower meatal fossa on
frontal and nasal with lateral supraorbital wall;
no diastemata between incisors; broader PI with
strong lingual cusp; stronger protoloph on P2;
and no or only minimal contact between first and
fourth metatarsals. Additionally they differ from I.
(. ii, '.u.oirus) by having well defined meatal fossa
on frontal and nasal; relatively shorter braincase;
and external auditory meatus not closed ventrally by
converging postglenoid and paroccipital processes.
Occurrence.-Early to latest Pleistocene of
the central and eastern United States.
Remarks.-Each of the four extant species of
i'q,/,,, is the type species of a named subgenus:
T. (Tapirus) terrestris; T. (Pinchacus) ih,._/../../..,
T. (7T-qi-, /!) bairdii; and 7. (Acrocodia) indicus
(Hershkovitz 1954). Matthew and Granger (1923)
named a new subgenus and species of Japirus, T.
(-i ..*:.. jidrus) ,,in,:_i i./ based on late Pleistocene
fv.,-- from China. Starting with Colbert and
Hooijer (1953), 13. :.'i. 'ii ,, has generally been
used at full generic rank, e.g., McKenna and Bell
(1997); Tong et al. (2002), Wu et al. (2006), Louys
et al (2007). However, Tong et al. (2002) and Tong
(2005) hypothesized that ....//o ,'rus i;.; .,"
evolved in China from within Tapirus (Colbert
& Hooijer 1953:90 also implied this is the case),
such that its use at the generic level makes 7.'q~ ,,.
paraphyletic. For that reason I treat it here at its
original rank, a subgenus of Tapirus.
Simpson (1945) and Ray and Sanders (1984)
admonished against formally naming subgenera
for North American fossil species of Tq .i'rus. But


with the greatly increased number of known fossil
species and a better understanding of their cranial
morphology and phylogenetic relationships, the
clade containing Tapirus veroensis and two other
species is clearly distinct from those represented
by the four modem subgenera (Hulbert & Wallace
2005; also see phylogeny section below) and
deserving of a formal name. Other than those three
species, 7,',ii,, polkensis (Olsen) is the most
likely to also be included in T. (H:./.._. .; '..), but
its phylogenetic relationships with T. (7:,i'q /i)
and T. (H.. /......./;.i veroensis are equivocal at
present (see below), so it is not assigned to the
new subgenus. Tapirus webbi Hulbert and Tapirus
johnsoni Schultz, Martin, and Corner are clearly
not members of T. (H?. i ./.. i..',/; , and the same
is likely also true of Kapirus simpsoni Schultz,
Martin, and Corner, Ty iii n- merriami Frick, and
Tapirus, .pi j o; , Merriam, butthey are currently
too poorly known to resolve their relationships.
All of the relatively well known fossil species
of Ti iP ~.. from South America (e.g., Tapirus
mesopotamicus), China (e.g., Tapirus hezhengensis,
Tapirus sanyuanensis), and Europe (e.g., Tapirus
arvernensis, 7T' i i ,,, ), J ,. .i) differ from T.
(Helicolapirus) in various combinations of cranial
and dental characters (based on descriptions in
Boeuf 1991; Rustioni 1992; Tong et al. 2002;Tong
2005; Ferrero & Noriega 2007; Deng et al. 2008),
and none appear to be referable to the new subgenus.
However, an inclusive phylogenetic analysis will
be needed to determine this more securely. Given
that the southeastern U.S. was always the primary
stronghold of T. (H. .-... ../ ; '., it is logical to look
for its origins in Neogene tapirs from this region.
In addition to the already mentioned T. polkensis,
other possibilities are the specimens from the late
Miocene Withlacoochee River 4A site in Florida
described by Hulbert (2005) or the larger, unnamed
species of tapir from the early Pliocene Palmetto
Fauna (Webb et al. 2008:fig. 6B). Unfortunately
neither is known from relatively complete skulls,
such that their phylogenetic relationships can be
determined for the most part only on the basis of
small samples of teeth. For tapirids, that rarely
produces satisfactory or conclusive results.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ",, )


TAPIRUS (HELICOTAPIRUS) LUNDELIUSI
SP. NOV.
Figures 3-18
Tapirus sp. Webb 1974, table 2.1 (in part); Webb and Wilkins
1984, table 1.
Tapirus n. sp. L.- ..ii. and Hulbert 1995, p. 68 and table 2;
Hulbert 1999, p. 53A.
,m ," late Blancan p*.. c;.- of Tapirus. Hulbert '' pp.
467, 474, 482 & 489, figs. 5D, 12B.
Holotype.-UF 160715, partial young
adult skeleton that includes cranium, mandible,
vertebrae, ribs, sternebrae, and fore- and hind limb
elements. GC-l....,.J 12-19 December 1994 by A.
E. Pratt, R. Toomey, and R. C. Hulbert. Found in
near full articulation.
Paratype.-UF 121736, partial subadult
skeleton that includes cranium, mandible, vertebrae,
ribs, and fore- and hind limb elements. Collected in
1989 and 1990 by R. C. Hulbert and G. S. Morgan.
Etymology.-The species name honors
Ernest L. Lundelius of the University of Texas
at Austin for his numerous contributions to the
study of Pleistocene mammals, including Tapirus
(Lundelius 1972; Lundelius & Slaughter 1976).
Type Locality.-Haile 7C (see Appendix 1).
Occurrence.-Very early Pleistocene (late
Blancan, ca. 1.6-2.6 Ma) of peninsular Florida.
Possibly also late Blancan of North Carolina.
Referred Specimens.-Haile 7C: in addition
to the holotype and paratype, this locality produced
four other skeletons with skulls and mandibles,
two young juveniles (UF 162351 and 206877), and
two young adults (UF 206876 and 206878), and
portions of p-.t, r ,i.,1 skeletons from four other
individuals (UF 149299, 160713, 162352, and
177735). Also the following isolated elements: UF
212272, Ml; 212273, P3; 212274, II; 212275, il;
177736, tibia; 177737, scapula; 177738, patella;
177739, metatarsal 2.
Haile 7G: Very young juveniles: UF 224663,
224664,224668,224676,243306,244089,244513,
247104. Young juveniles: UF 224666, 224673,
224681,2 i-'-0 1, 244090,244512,244514,244515,
244516, 247101, 247103. Juveniles: UF 223827,
224662, 224665, 244519, 247102. Subadults:
UF 224671, 224682, 224684. Young adults: UF


221720, 224672,224679, 224685,224686,244503,
244511. Full adults: UF 224670, 224674, 224680,
224683, 244502, 244504. Old adult: UF 244600.
Santa Fe River 8: UF 177509, squamosal;
15095, M2 and M..s,; 177512, 12; 177513, i3;
177510, axis; 177511, radius; 177514-177516,
unciforms; 240325, trapezoid; 177520, scaphoid;
177521, pisiform; 177517-177519, magnums;
177522-177523, cuneiforms; 15096, 177524
metacarpal 3s; 177525, metacarpal 4; 17527-
177528, metacarpal 5s; 15098, 177531-177532
tibiae; 177533, fibula; 177534-177535,
cuboids; 240382, navicular; 177536-177538,
ectocuneiforms; 177526, metatarsal 2; 177529,
metatarsal 3; 177530, metatarsal 4; 15097, 18476,
177539-177544, phalanges.
Santa Fe River 8A: UF 177546, frontal;
177547, fused supraoccipital, interparietal, and
parietal; 16763, associated(?) P1 and p2; 14261,
M3; 177548-177550, thoracic vertebrae; 177717-
177718, lunars; 45727, pisiform; 177716, uniform;
177719, metacarpal 2; 177720-177721, metacarpal
4s; 14259, 177715 astragali; 14524, calcaneum;
177712-177713, cuboids; 177714, ectocuneiform;
45728, navicular; 177724, 240225, metatarsal 2s;
14257, metatarsal 3; 177722-177723, metatarsal
4s; 14260, 177725-177731, phalanges.
Santa Fe River 8C: UF 177501, scapula;
177502, ulna; 177503, metacarpal 3; 177504,
metacarpal 4; 177505, fibula; 177506, astragalus;
177507, calcaneum; 177508, metatarsal 4.
Withlacoochee River lA: UF 24-'.', M3;
241366, metatarsal 4.
Waccasassa River 9A: UF 240682, lower
molar.
Inglis lA: UF/TRO 2625 maxilla with Ml-
M3; UF 66134, nasal; 66125, supraoccipital; 18175,
maxilla with M1-M2; 115970, maxilla with P1-P4;
18176, Il; 18177, DP4; 115969, p3; 176712, dp3
or dp4; 176713-176714, two lower cheek tooth
fragments; 45305, mandibular symphysis; 22689,
177701, axis vertebrae; 177702-177703, cervical
vertebrae; 177704-177709, thoracic vertebrae;
18178, humerus; 45309-45310, 45447, 177835,
distal humeri; 177710, lunar; 244473-244474,
scaphoids; 244475-244477, magnums; 244478,






HULBERT JR.: Blancan Tapirus from Florida


uniform; 244479, pisiform; 176715, metacarpal 3;
45549, metacarpal 4; 115971-115975, metacarpal
5s; 45278, distal femur; 45306-45308, tibiae;
45440, astragalus; 18179, 115978, calcanea;
177711, cuboid; 244480-244481, ectocuneiforms;
22690, fused mesocuneiform, ectocuneiform, and
metatarsal 1; 45450, 115977, 176701, metatarsal
2s; 18180, 115976 metatarsal 3s; 45448, metatarsal
4; 18181-18183, 45441-45445, 176709-176711,
phalanges.
Inglis 1B: UF 97116, P2; 97117, mandible
fragment with m3; 97118, p3; 97119, i2; 97120,
metacarpal 3; 97121, cuneiform.
Inglis ID: UF 224247, humerus; 224248,
metacarpal 2; 224249, femur; 224250, thoracic
vertebra.
Walden Pit 2: UF/TRO 1476, mandible
fragment with m3.
Diagnosis.-Species of T. (1i. hi. .,., ,i,-,
significantly smaller than Tapirus haysii; smaller on
average than }apirus veroensis, but with extensive
overlap in most measured characters. Sagittal crest
slightly more pronounced than in T veroensis and
T. haysii. Interparietal fused to occipital prior to
eruption of M2 (fusion in 17 veroensis and T. haysii
occurs after eruption of M2, sometimes not until
after eruption of A.. I). Only a single large lacrimal
foramen present medial to the posterior process of
lacrimal; not visible in lateral view (T7 veroensis
has two large lacrimal foramina and usually at
least one is visible in lateral view). Posterolateral
margin of nasal not sharply downturned (present in
T7 haysii). Dorsal table of frontal relatively smaller
than in T. veroensis and T. haysii, with an irregular
surface (not smoothly inflated as in T1 veroensis
and T. haysii). Small dorsal flange on maxilla
medial to posterior process of premaxilla (absent
in T. veroensis and T. hJi .--). P1 lacks a strong
transverse loph, but its width is relatively large.
TAPIRUS (HELICOTAPIRUS) HA YSII LEIDY
1859
Figure 2
Referred Blancan Specimens.-Santa Fe
River 1: UF 177841, partial M2 or ..fl; 177842,
associated right and left p4s. Santa Fe River 1B:


UF 213920, 224631 Mls (?). Kissimmee River
2: UF 51250, P3. Kissimmee River 6: UF 52601,
metacarpal 5. Inglis IC: UF 177844, Pl. De Soto
Shell Pit 5: UF 223920, associated right maxilla
with DP2-DP4 and left maxilla with DP1-DP4;
232075, cuboid. Devil's Elbow 2: UF 177843, M3.
See Appendix 1 for locality information.
Identification.-These specimens (Fig. 2;
Table 2) are much larger than any found in the
Haile 7C and 7G samples of Tl ,wi., lundeliusi,
and instead fall within or near the OR of a7pirus
haysii (= Tq ,ii -. copei) from the Irvingtonian of
Florida (Ray 1964; Hulbert 1995) and Pennsylvania
(Simpson 1945). They are also similar in size to the
holotype and referred specimens of Japirus haysii
from North Carolina (Ray and Sanders 1984). The
matching p4s (UF 177842) have the greatest known
L and AW for that tooth among all specimens
assigned to T7 haysii (if western specimens of a
large tapir represent a different species, Tapirus
merriami, as proposed by Jefferson 1989, in
contrast to Ray and Sanders 1984). The reference
of the Florida Blancan sample to T. haysii is also
supported by a transverse loph on the Pl, UF
177844 (Fig. 2F), a characteristic feature of this
species, and one rarely present in T. hlundeliusi or
17 veroensis. UF 213920 and 224631 are isolated
upper molars that are either relatively large Mis or
small M\ As the other Santa Fe River specimens
are of relatively large size, notably UF 188742, an
assignment as Mls is considered more likely. The
dimensions of the metacarpal 5 all fall within the
OR of the early Irvingtonian sample of 7' haysii
from Florida provided by Hulbert (1995), and
are significantly larger than those observed in I.
hlundeliusi.
Chronologic Distribution.-The range of
Tapirus haysii in Florida is emended from that in
Hulbert (1995) to be early late Blancan through
early Irvingtonian. The lone middle Irvingtonian
record of I7 haysii in Florida listed by Hulbert
(1995) was McLeod Rock Pit, based on two
juvenile specimens. Increased numbers of juveniles
of Tapirus veroensis no longer support placing the
McLeod fossils in T7 haysii; they are instead more
likely to be early representatives of T veroensis.







82 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)






HULBERT JR.: Blancan Tapirus from Florida


D















5cm









5 cm















Figure 4. Skull of holotype of Tapirus lundeliusi sp. nov., UF 160715, from Haile 7C. A, dorsal view; B,
ventral view; C, occlusal view of right P1-M3; D, right lateral view; and E, posterior view zygomaticc
arches removed). This skull was restored so that a cast could be made for public display at the Florida
Museum of Natural History.







BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)


A


'C


5 cm


C-
A~.. tr Al


S.a-


3 cm


'.-TV


w^.,.






HULBERT JR.: Blancan Tapirus from Florida


5 cm


4,
4

I
-


4'


A
#

~: ( ~


/






I~df


Figure 5. Skull of paratype of Tapirus lundeliusi sp. nov., UF 121736, from Haile 7C. A, dorsal view;
B, ventral view; C, occlusal view of muzzle with right and left 11-13, C. D, left lateral view (reversed); and
E, posterior view zygomaticc arches removed). The M2 is not yet fully erupted in this individual.


3 cm


I






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. f',, )


TAPIRUS SUBGENUS AND SPECIES
INDETERMINATE
Figure 19

Referred Specimens.-US 19 Bridge Site: UF
255357, partial juvenile maxilla with DP3 (missing
metacone); UF 255358, mandibular fragment with
dp3 (missing most enamel on labial side of crown);
UF 248788, talonid of p2; UF 247180, navicular;
UF 248789, proximal phalanx of digit 2 or 4;
Hubbard Pit: UF 249089, dp3; Kissimmee River
6: UF 51249, p2; Sommers Pit: UF/TRO 1469,
unworn p4 (crown only, missing roots); Walden
Pit 1: UF/TRO 1489, mandibular fragment with
posterior root of p2 and moderately worn p3. See


Appendix 1 for more information on localities.
Measurements.-UF 255357: AW, ca. 18.3
mm. UF 255358: L, 19.8 mm. UF 249089: L, 19.5
mm; AW, ca. 12.8 mm; PW, 13.0 mm. UF 51249:
L, 20.0 mm; PW, 12.4 mm. UF 248788: PW, 12.4
mm. UF/TRO 1469: L, 20.2 mm; AW, 15.1 mm;
PW, 16.7 mm. UF/TRO 1489: L, 17.9 mm; AW,
13.0 mm; PW, 14.2 mm. UF 247180: greatest W
of astragalar facet, 29.8 mm; greatest depth of
astragalar facet, approximately 21 mm. UF 248789:
greatest L, 24.7 mm; proximal W, 16.4 mm; distal
W, 13.5 mm.
Tooth Identification.-The sole tooth in UF
255387 is definitely deciduous, as the unerupted
permanent premolar is visible in the crypt medial


Table 2. Comparison of dental measurements of late Blancan specimens of 7./,1/ ',o haysii from Florida
with samples from the early Irvingtonian of Florida and the middle Irvingtonian Port Kennedy Cave Site
in Pennsylvania (MIN-MAX (N)). Values for the latter after Simpson (1945).


Catalogue
Number
UF 223920
UF 223920
UF 223920
UF 177844
UF 177844
UF 51250
UF 51250
UF 51250
UF 213920
UF 213920
UF 213920
UF 224631
UF 224631
UF 224631
UF 177841
UF 177843
UF 177843
UF 177843
UF 177842
UF 177842
UF 177842


Tooth
Locus
DP4
DP4
DP4
P1
P1
P3
P3
P3
M1
M1
M1
M1
M1
M1
M2/M3
M3
M3
M3
p4
p4
p4


Measure-
ment
L
AW
PW
L
PW
L
AW
PW
L
AW
PW
L
AW
PW
AW
L
AW
PW
L
AW
PW


Value
(mm)
23.6
26.7
23.2
20.7
19.6
24.0
27.5
26.6
25.5
31.7
28.2
26.3
30.8
28.0
32.4
29.9
32.5
26.8
28.8
22.8
22.7


Irvingtonian
of Florida
22.3-26.1 (5)
24.9-30.3 (5)
22.8-27.0 (5)
19.7-23.3 (7)
17.3-22.0 (7)
21.6-24.5 (10)
25.7-29.0 (10)
25.4-29.3 (11)
23.9-25.9 (12)
26.9-30.7 (11)
24.5-28.5 (12)
23.9-25.9 (12)
26.9-30.7 (11)
24.5-28.5 (12)
30.8-34.5 (19)
27.1-29.1 (6)
31.5-33.6 (6)
25.1-28.9 (6)
22.5-25.7 (6)
19.9-22.0 (7)
20.2-24.0 (6)


Port Kennedy
Cave, PA


22.4-
19.6-
22.7-
27.0-
26.1-
25.8-
28.9-
25.8-
25.8-
28.9-
25.8-
31.0-
26.8-
31.0-
26.5-
24.1-
18.3-
19.5-


-24.9 (3)
-21.5 (3)
-24.5 (5)
-29.5 (4)
-29.0 (4)
-26.4 (5)
-31.1 (5)
-27.9 (4)
-26.4 (5)
-31.1 (5)
-27.9 (4)
-34.9 (9)
-29.2 (5)
-34.1 (5)
-29.0 (5)
-24.9 (6)
-21.7 (8)
-22.8 (8)






HULBERT JR.: Blancan Thpirus from Florida


to the labial root. An identification as a DP3 is
based on its relative position to the IOF in similar
age individuals, the orientation and height of the
protoloph, and strength of the parastyle. UF/TRO
1469 is identified as a p4 on the basis ofan AW/PW
ratio of about 90 percent (which eliminates molars)
and absence of a metalophid (which eliminates
p2 and p3). In contrast, the complete tooth in UF/
TRO 1489 has a metalophid and a small portion
of the mental foramen is present ventral to the set
of alveoli anterior to the tooth. This combination
of features is only t.-. -<-i in the p3 of Tapirus.
UF 249089 and 255358 are identified as dp3s
on the basis of the presence of metastylids (not
found on permanent teeth), relatively narrower but
otherwise molariform morphology, and low crown
height. They compare very favorably to the dp3 of
UF 223991. An identification of either as a dp4 is
regarded as possible but less likely because of their
relatively low hypsodonty.
DESCRIPTION OF TAPIRUS LUNDELIUSI
CRANIUM
Cranial measurements are listed in Table
3. The rostrum is narrow with a relatively short
postcanine diastema (Figs. 4-6). In lateral view, the
posterior process of the premaxilla terminates in
an acute point approximately dorsal to the anterior
margin of the PI; the maxilla is exposed medial to
the posterior process of the premaxilla and there
forms a slight (UF 160715, 224680, 224682) to
modest (UF 121736, 206878, 244502), ca. 30 mm
long, vertical flange for embrasure of the nasal
septum (Figs. 4D, 5D, 6-7); and the dorsomedial
border of the maxilla thins posterior to the maxillary
flange and projects medially (but is not rolled
ventrally). The narial incision terminates dorsal
to the middle or posterior half of the orbit (Figs.
4D, 6A). The posterodorsal ascending process of
the maxilla forms the margin of the narial aperture
back to the middle of the orbit where it articulates
with the descending process of the nasal; this
portion of the maxilla is widely exposed dorsally,
not twisted laterally (Figs. 4A, 5A). The anterior
supraorbital process of the frontal overlaps the
lacrimal dorsally and articulates medially with


the maxilla and descending process of the nasal.
Dorsal to the orbit, the frontal forms a broad,
concave shelf that carried the meatal diverticulum;
posterior to the orbit, this groove on the frontal
narrows, curves dorsomedially, and becomes
shallower, with a poorly demarcated posterior
margin. As the groove approaches the midline of
the skull it curves anterolaterally, and becomes
deeper and broader as it reaches the nasal. The
anterior border of the "spiral" groove on the nasal
is well defined on all subadult and adult individuals
(N = 9) and even on the juvenile UF 162351 (Figs.
8-9). The groove finally ends at the lateral side of
the nasal near its widest point. The shape and size
of the preserved nasals vary, but in all subadults
and adults the lateral edge is posteriorly convex
and anteriorly slightly concave (N = 8) or straight
(N = 1), they are slightly arched anteroposteriorly,
and thin dorsoventrally (Fig. 6A). Nasal length
is equal to or greater than twice maximum nasal
width in 8 of 9 subadult and adults. In the one
very young juvenile with --..i preserved nasals
(UF 247104, Fig. 9), the meatal fossa is faint and
the anterior portions of the nasal are not ossified,
such that there is no osteological contact along the
midline of the right and left nasals (there may have
been cartilaginous contact). A second very young
juvenile, UF 224668, appears to have the same
nasal morphology as UF 247104, although both
nasals are incomplete such that their lack of contact
can not be ascertained unambiguously. Slightly
older juveniles, such as UF 162351, have nasals
with an overall morphology resembling those of
adults, including a long region of contact along the
midline.
The dorsal table of the frontals (the
region bounded anteriorly by the nasals and
posterolaterally by the temporal crests) is relatively
small because the parasagittal ridges are strongly
concave and meet anterior to the frontoparietal
suture (Fig. 8). The dorsal surface of the frontals
is not flat; there is a pronounced medial ridge on
the midline, and a smaller and slightly lower but
notably raised region on the parasagittal ridge
directly p-rtei-,, to the descending process of
the nasal. Internally, the frontals contain sinuses









Table 3. Measurements (in mm) on skulls of Tapirus lundeliusi n. sp. from the early Pleistocene of Florida. For comparison, values for
other species of Tapirus are listed at base of table (when N>1, first line is xs, second line is OR(N), and third line is percentage difference
of x with mean value of lu hndeliusi). Occipital HT measured from ventral surface of basioccipital to dorsal surface of supraoccipital.


Specimen
Number
UF 160715
UF 121736
UF 115970
UF 206876
UF 206878
UF 224672
UF 224674
UF 224679
UF 224680
UF -4 I
UFY2 ,-,
mean s
CV
T. johnsoni


Basal L


P1-P4 L


75.5
78.4
75.45
75.2
77.6

78.4

73.1


329.74.51
1.37
353(1)


108%
T. simpsoni -

T polkensis 282.78.74
273-290(3)
86%
T haysii -


T veroensis 353.512.02
345-362(2)

7T terrestris 350.615.83
328-384(16)

T. bairdii 379.511.15
361--' i:19)
115%
T indicus 403.34.37
399-407(3)
122%


80.1
76.72.27
2.97
72.74.22
67.2-77.1(4)
95%
84.9(1)
111%
60.33.52
.2 .-. '(19)
79%
87.20.70
86.7-87.7(2)
114%
77.93.92
7- -.. -4 '(12)
102%
71.83.42
66.0-80.8(18)
94%
74.93.23
67.2-80.9(21)
98%
89.04.22
84.5-94.7(4)
I .'1


M1-M3 L


69.9


67.7
71.3

70.3

66.05
73.6
72.2
70.22.59
3.70
66.22.28
63.6-i.: ,-
94%
73.8(1)
S-I ..

55.62.43
52.8-62.4(15)
79%
80.01.47
77.5-81.3(5)

70.04.33
63.6-77.9(11)

65.42.48
61.0-71.4(18)
93%0
65.52.53
61.1-7( \: )
93%0
78.1+6.52
70.6-82.4(3)
111%


PI-M3 L


144.3


142.5
147.8

147.65

138.6

152.3
145.54.79
3.29
137.29.44
130.5-143.9(2)

159.5(1)
110%
115.66.17
108.2-127.8(9)
79%
166.30.41
166.0-166.6(2)
1 4 ..
1-- ..78.92
133.4-159.0(6)
99%
134.84.93
125.3-147.4(18)
93%
138.15.14
126.0-148.4(19)

164.011.06
151.8-173.3(3)
113%


C-PI DL


Muzzle W


Occipital HT


94.5
93.3

101.6


107.5


37.54.82
12.88
44.44.34
40.4-49.0(3)
118%
43.9(1)
117%
33.5+4.32
26.6-41.9(18)
89%
48.52.55
45.6-50.4(3)
129%
40.76.52
30.0-50.1(11)
108%
47.94.35
40.7-57.4(17)
128%
54.04.41
43.5-60.7-. ..'?)
144%
53.09.31
40.1-62.4(4)
141%


44.01.07
2.43
43.2(1)

98%


39.21.76
36.0-41.4(7)
89%
52.33.32
48.7-55. '-1i
119%
48.03.42
5.2-52.5(6)
109%
43.03.37
37.0-50.0(16)
98%
50.62.94
44.6-55.1 .',
115%
50.81.14
49.5-52.2 I-4
I ..


98.0
99.05.76
5.82
103.0(1)

i- I-'


82.92.17
80.6-86.1(6)
84%
114.42.92
111.7-117.5(3)
115%
I .. ;3.43
105.7- 2- :(1
109%
122.85.99
115.5-132.2(16)
124%
120.35.12
110.0-132.7(21)
122%
115.76.59
106.5-T 11' i. 4. i
117%


Occipital
Condyle W
74.1
71.3

78.0



81.5
79.6

86.8
78.55.49
6.98
77.6(1)

99%


65.11.69
63.7-67.4(4)
83%
83.01.29
81.5-84.5(4)
106%
81.76.69
76.1-89.1(3)
104%
71.43.88
64.7-79.3(16)
91%
83.44.10
77.1-91.9(19)

88.73.47
86.9-93.9(4)
113%






HULBERT JR.: Blancan Tapirus from Florida


A narial incision



iof



max. flange i


K" ,/ -\ '


asc. proc. max.%


3 cm


Figure 6. Left lateral views of anterior portions of skulls of Tapirus lundeliusi sp. nov. from Haile 7G. A,
UF 224682, subadult with C, P1-P3, DP4, Ml, and erupting M2; B, UF 224680, full adult with 12, Pl-
M3. Abbreviations: asc. proc., ascending process; iof, infraorbital foramen; max., maxilla; ppl, posterior
process of lacrimal.


3 cm


% ppl


-'*-V.I


*^ ../-





BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)


T6,J


3 cm


iof

B


Figure 7. Left lateral views of anterior portions of skulls of Tapirus lundeliusi sp. nov. from Haile 7G.
A, UF 247101, very young juvenile with DP1-DP4; B, UF 224662, juvenile with DPl-M1. Abbreviations
as in Figure 6.





HULBERT JR.: Blancan Tapirus from Florida


A

























5 cm



Figure 8. Dorsal views of nasals, frontals, and sagittal crests in adults of Tapirus lundeliusi sp. nov.
A, UF 160715 from Haile 7C; B, UF 224680 from Haile 7G. Note variation in length of nasals.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)


which begin forming in young juveniles. The
posteriorly converging parasagittal ridges meet to
form the sagittal crest in the posterior portion of
the frontal and it continues onto the parietals (Fig.
8). The ontogenetic sequence of skulls from Haile
7C and 7G reveals the developmental pattern of
the sagittal crest (Figs. 8-10). Very young juveniles
have smoothly rounded parietals without any
indication of parasagittal ridges (UF 224664) or
very weak ridges far from the midline (UF 224668,
247104). This condition persists into the first part
of the young juvenile stage (UF 162351), but
the ridges thicken and move towards the midline
later in this stage, about the same time as the Ml
begins to erupt (UF 206877, 244514, 247103).
The parasagittal ridges become narrower and taller


to form a sagittal crest consisting of two distinct
ridges separated by a narrow groove in juveniles
and subadults (UF 121736, 224682). In young and
full adults (UF 160715, 206876, 206878, 224680),
this groove is filled to form a single sagittal crest,
and the crest is slightly taller than in the subadults.
In lateral view, the dorsal profile of the sagittal
crest is slightly arched, reaching an acme only
several millimeters higher than the dorsal tables
of the frontals or occipital (Figs. 4D, 5D). The
parasagittal ridges begin to diverge posterolaterally
about 40 to 50 mm from the posterior edge of the
skull, forming the anterolateral edges of a flat,
triangular-shaped region that is continuous with
the dorsal surface of the sagittal crest. Much of this
flattened region originally ossified as a distinct,





meatal fossa
on frontal









1e
'ALI '


Figure 9. Dorsal view of the braincase in a very young juvenile of Tapirus lundeliusi sp. nov., UF 247104
from Haile 7G.






HULBERT JR.: Blancan Tapirus from Florida


WO.r -


*~.AtUV

WA


Figure 10. Dorsal views of the skulls of Tapirus lundeliusi sp. nov. A, juvenile UF 224662 from Haile 7G;
B, subadult UF 121736 from Haile 7C. Contrast the development of the sagittal crest and the fusion of
interparietal and occipital bones in these individuals to the conditions in the very young juvenile in Figure
9. Scale bars are 5 cm in length.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)


large, triangular interparietal bone (observable in
UF 162351, 206877, 224664, 224668, 244514, and
247104). The interparietal fuses with the occipital
as early as the young juvenile stage (UF 224666)
but more often in the juvenile stage (UF 224662).
Fusion of the interparietal with the parietals starts
in the subadult stage (UF 224682) and is completed
by the young adult stage (UF 160715, 206876).
The lambdoidal crests are strongly developed and
project primarily posteriorly in young juveniles
(Fig. 9), but posterolaterally in subadults and adults
(Figs 4A, 10B).
In lateral view, the infraorbital foramen
is located dorsal to the DP2 in juveniles (Fig. 7;
N = 8) and the P3 in subadults and adults (Fig.
6; N = 11); its posterior margin is formed by a
very thin (ca. 5-mm-wide) strut of the maxilla.
The lacrimal bone articulates anteriorly with this


maxillary strut, dorsally with the ascending process
of the maxilla, and ventrally with the jugal (Fig.
7A). Posteriorly, the lacrimal forms the anterior
margin of the orbit. The dorsoventral height of
the lacrimal greatly exceeds its anteroposterior
length, and it has a generally smooth, moderately
concave lateral surface (Figs. 5D, 6-7). A distinct
anterior lacrimal process is usually absent (N = 8)
or merely a small rugosity (N = 2). All specimens
in which the bone is preserved have a short, broad,
flat posterior lacrimal process. Directly medial to
the posterior lacrimal process (and not visible in
lateral view; Figs. 6, 11) is a single, large lacrimal
foramen (N = 12). On UF 121736, it is 7.5 mm tall
and 4 mm wide. There are no other large lacrimal
foramina, but there are three to five much smaller
ones located dorsal and lateral to the larger opening
(Fig. 11A).


1 cm


Figure 11. Condition of the lacrimal foramen in Tapirus lundeliusi sp. nov. A, UF 224673, isolated left
lacrimal of young juvenile in posterior view; B, UF 224680, left rostrum of adult in posterolateral view,
arrow points to lacrimal foramen. Every individual of this species in which the lacrimal is preserved,
regardless of ontogenetic age, has a single large lacrimal foramen that is not visible in lateral view.






HULBERT JR.: Blancan Tapirus from Florida


The lateral opening for the external auditory
meatus is circular, ca. 15 to 17 mm in diameter; the
meatus is partially closed ventrally by a posterior
proj section of the postglenoid process and an anterior
projection of the mastoid process, but the two do
not connect (Fig. 4D). An anterc. i ..1i- directed
ridge branches off the descending lambdoidal crest
about halfway through its length and continues
down to the end of the mastoid process of the
squamosal. This ridge is more pronounced in
adults than subadults. The mastoid process of the
petrosal is visible wedged between this ridge and
the lambdoidal crest. On UF 160715, the mastoid
process is oval, about 16 mm wide and 28 mm tall,
and faces more posteriorly than it does laterally.
On UF 121736, it is slightly narrower and faces
posteriorly and laterally in about equal measure. The
stylomastoid foramen opens at the ventral suture
between the mastoid and squamosal into a shallow
groove that runs ventrally down the squamosal just
anterior to its suture with the paroccipital process.
The long paroccipital process projects much more
ventrally than either the postglenoid process or the
occipital condyles (Figs. 4D-E, 5D-E). The back of
the skull dorsal to the foramen magnum is oriented
nearly perfectly vertically.
The incisive foramen is long, extending from
just p.-.- i$-,.rto the 13 to the level of PI (Fig. 4B).
In most subadults and young adults the incisive
foramen ends lateral to the anterior half of the P1,
while in full adults it ends lateral to the posterior half
of the PI or even level with the Pl/P2 contact. The
hard palate formed by the maxillae and palatines is
moderately arched transversely and bears several
foramina, the largest of which opens anteriorly
into a shallow sulcus at about the same level as the
anterior border of the Ml. The palate ends along
the midline at a point on line with the middle of the
M2 with a moderate rugosity in adults (UF 160715,
206876, 206878). The choane (internal nares) are
relatively narrow. The medial wall of the orbit is
not well preserved in any available specimen. The
basicranial region is dominated by a very large
lacerate foramen lateral to the basioccipital and
posterior to the alisphenoid and postglenoid region
of the squamosal (Figs. 4B, 5B). The smaller


hypoglossal foramen is located near the pl, teri;,
margin of the lacerate foramen and is approximately
circular with a diameter of 6-9 mm in adults. The
width of the foramen magnum is slightly greater
than its height. Minimum ventral width between
the occipital condyles in adults is about 12 percent
of occipital condyle width (N = 3).
MANDIBLE
The ventral border of the ramus is gently
curved posterior to the uplifted symphysis (Fig.
12). There are no diastemata between any of the
incisors or canine, and the postcanine diastema
is relatively short (Table 4). The symphysis has
a deeply concave dorsal surface. Posteriorly,
the angle is strongly inflected medially and its
margin is thick and well developed (Fig. 13). The
posteromedial region of the mandible is concave
and in subadults and adults bears pronounced ridges
that start at the thickened posterior margin. The
large mandibular foramen opens posterodorsally
and is located at the same height as the cheekteeth
(Fig. 14). The mediolateral axis of the mandibular
condyle is oblique, with the medial side lower than
the lateral side. Dorsoventral height from the angle
to the mandibular condyle is relatively greater than
in other species of aipirus (Table 4). The anterior
margin of the ascending ramus projects anteriorly
in lateral view (Figs. 12D, 14B) so that at the
level of the condyle it partially overlies the m3.
The coronoid process is tall, extending 40-50 mm
beyond the height of the condyle. The masseteric
fossa is large and deep, with its ventral border
poorly defined but located at about the level of the
crowns of the cheekteeth. The mental foramen is
located ventral to the dp2 or p2 (N = 27). One rare
variant is UF 244502 with two mental foramina,
one ventral to the p2 and another ventral to the p3.
This unusual feature is found on both the right and
left dentaries of this individual.
DENTITION
The incisors and canines are morphologically
like those of other New World Tapirus, in particular
there is a reduced upper canine, a caniniform 13,
an enlarged, -.j., bi.-(e il, and a greatly reduced
i3 (Figs. 5C, 12A, C, 14A). The P1 is slightly









Table 4. Statistics of measurements (in mm) on mandibles of Iapirus himdeliusi sp. nov. from the early Pleistocene of Florida and those of


other species of 7, -. When N>1, first line is xs, second line is OR(N). For T.
third line is percentage difference of its mean value with that of T1 lundeliusi.


lundeliusi, third line is CV; for other species of Tapirus,


Condylar L p2-p4 L


ml-m3 L p2-m3 L


c-p2 DL


Muzzle W H at p2


T lundeliusi 283.98.70
278-290(2)
3.06

T polkensis 243.27.76
237-255(5)
86%

T johnsoni 298.09.54
288-307(3)
105%

T webbi 329.33.89
326.5-332(2)
116%

T havsii 343.613.54
326-359(4)
121%

T veroensis 308(1)

109%

I terrestris 286.715.31
267-315(16)
101%

T bairdii 304.09.60
289- '.12 ,


T indicus 336.83.88
332.5-3 i )
119%)


63.62.06
59.9-66.1(8)
3.24

50.02.33
46.8-53.8(16)
79%

59.72.44
55.8-63.0(13)
14

65.61.87
62.1-68.3(8)
103%

72.93.50
67.7-78.5(6)
115%

63.933.06
59.1-67.1(6)
100%

60.43.16
55.9-67.1(16)
95%

62.02.56
58.3-67.9(19)


74.43.70
69.5-77.6(4)
117%


72.62.98
69.1-77.6(8)
4.03

57.523.83
52.4-61.6(16)
79%

69.42.48
66.0-73.6(12)
96%

76.62.42
72.2-79.4(6)
105%

82.84.83
76.7-89.6(6)
116%

72.03.02
65.9-76.8(22)
99%

67.62.78
63.0-74.8(16)
93%

66.82.33
61.7-70. 'i' ,
92%

81.17.43
70.4-87, i
112%


135.54.82
1295.-I- -. : ,,
3.56

107.24.14
101.2-114.4(11)
79%

128.64.41
122.0-136.0(11)


143.14.14
137.5-147.5(4)
106%

158.16.13
153.4-166.1(5)
117%

134.20.39
133.9-134.7(3)
99%

127.25.76
118.5-141.1(15)
94%

128.24.74
119.7-138.5(18)


155.212.31
137.9-165.3(4)
115%


45.53.53
38.5-49.1(11)
7.75

39.24.37
32.5-47.0(18)
86%

58.15.20
52.5-564.2(4)
128%

58.11.96
55.2-61.2(6)
128%

59.46.11
52.4-68.4(5)
131%

47.44.08
41.9-53.4(7)
104%

55.55.35
48.2-67.6(17)
122%

65.34.61
53.9-73.1(22)
144%

58.34.99
53.3-614 -.i4
128%


46.22.79
43.5-4 i .,6)
6.03

36.92.24
32.5-39.2(10)
80%

42.72.14
40.4-44.6(3)
92%

50.04.21
46.6-55.7(4)
108%

52.91.60
51.1-54.2(3)
114%

44.72.28
41.1-48.0(6)
97%

44.24.04
36.6-53.8(15)
96%

51.13.46
47.6-61..i21,
111%

55.85.97
51 ;-, :(4)
121%


42.02.18
39.2-44.6(9)
5.20

33.12.63
27.4-37.7(13)
79%

46.13.00
42.4-49.7(4)
110%

44.81.92
41.1-46.6(6)


52.65.15
45.5-58.2(7)
125%

44.32.91
42.1-49.8(6)


45.55.29
37.6-55.2(17)
108%

47.62.15
41.7-51.3(22)
113%

54.12.46
51.6-56.5(3)
129%


61.13.19
56.2-66.7(8)
5.22

44.83.10
38.3-49.8(14)
73%

56.81.98
53.9-60.8(9)
93%

56.33.52
50.0-61.7(11)
92%

71.64.15
67.3-7F ?2 )
118%

59.63.44
53.7-65.9(23)


51.93.64
46.9-57.4(16)
85%

55.23.09
51.3-64 : )
90%

60.42.15
58.1-62.3(3)
99%


151.310.34
139.7- ., 5)
6.84

112.43.08
109.0-116.5(6)
74%

136.92.34
I 7-14(I I ..
90%


128(1)

85%


173.02.65
171-176(3)
1 -"


131.14.20
128.2-1' :4)


123.36.76
111.2-133(12)
82%

133.55.73
123.5-144(15)
88%


155.5(1)

103%


H at m3


condyle H






HULBERT JR.: Blancan Tapirus from Florida


5 cm


Figure 12. Mandibles of Tapirus lundeliusi sp. nov. from Haile 7C. A & C, occlusal views of symphysis;
B & D, lateral views of left mandible. A-B, UF 121736, subadult paratype with il-12, c, p2-p3, dp4,
ml-m2; C-D, UF 160715, young adult holotype with i3, c, p2-m3. Scale bar of 5 cm on right applies to
both B and D.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)


longer than wide (Figs. 4C, 15A-C, 16C; Table
5); its morphology is best shown on the slightly
worn UF 16763, 121736, 224671, and 224682.
The labial portion of the PI crown is united in a
sharp ectoloph; the paracone and metacone are
situated close to each other; and the paracone is
the tallest cusp on the PI although neither it nor
the metacone extend much beyond the level of the
remainder of the ectoloph (Fig. 6A). A narrow, low,
posterior cingulum wraps around the back of the
P 1, connecting the ectoloph with the bulbous, main
lingual cusp. A slight crest extends labially from


'~ I -0


the lingual cusp towards the metacone in four of
seven unworn or slightly worn Pis, while the other
three lack any trace of a transverse loph. About half
of the Pis have a much smaller and lower cuspule
located anterior to the main lingual cusp, while
the other half lack such a structure. With wear, the
P1 ectoloph remains a sharp ridge (UF 160715,
115920, 206876, 224686, 224674), while the
lingual cusp wears flat and merges with the smaller
cuspule (if present).
The AW of the P2 is on average 91 percent
of PW (N = 18). The P2 metacone and paracone


r

7


A~.
,y I


4; -


0 cm


Figure 13. Posterior view of right and left mandibles of UF 160715, holotype of Tapirus lundeliusi sp.
nov.






HULBERT JR.: Blancan Tapirus from Florida


3 cm












C5 cm














Figure 14. Mandibles of Tapirus lundeliusi sp. nov. A, occlusal view of symphysis, and B, medial view of
right mandible (reversed) of UF 206878, adult mandible with il-i3, c, p2-m3 from Haile 7C; C, medial
view of UF 221720, young adult mandible with right c and left p2-m3 from Haile 7G. Arrow in B points
to location of mandibular foramen. 5 cm scale bar applies to B and C.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


are well separated, unlike those of the Pl, but in
the unworn state also barely project beyond the
level of the rest of the ectoloph (Figs 6A, 15E,
16C). The P2 has a small parastyle and anterior and
posterior cingula. The transverse lophs are oriented
obliquely, especially the weak protoloph that
extends from the protocone towards the parastyle
where it connects near the base of the ectoloph. The
metaloph is oriented more transversely than the
protoloph, and is taller, uniting about halfway along
the height of the ectoloph. As in the succeeding
teeth, the metaloph curves posteriorly just prior to
uniting with the metaloph, so that it connects with
the metacone.
The P3 (Figs. 4C, 15A-C, 16C; Table 5) is
more molariform than the P2 in that AW exceeds
PW in most individuals (14 of 16) and the
transverse lophs are not as oblique. Compared to
the P2, the protoloph is stronger and unites with the
paracone instead of the parastyle, and the parastyle
is larger. When unworn, the P3 ectoloph is sharply
depressed between the paracone and metacone.
The P4 continues these trends, basically being a
larger version of the P3 with a stronger parastyle
and taller cusps and lophs (Figs. 4C, 15A-C). The
M1 through M3 also have large parastyles that are
separated from the paracones by deep creases; AW
greatly exceeds PW, especially on the M3. A small
cusp is variably present on the anterior side of the
protoloph close to the anterior cingulum. Small
lingual and labial cuspules are also variably present
between the protocone and hypocone, or paracone
and metacone, respectively, of the P3-M3. None of
the teeth have a cingulum around the labial base of
the metacone.
The DPI is approximately the same size
as the P1 (Table 7), slightly lower crowned, and
has a well formed but narrow transverse loph the
extends from the lingual cusp to the ectoloph (Fig.
16A). The DP2 is about the same length as the P2,
but much narrower, and has a stronger protoloph,
a much lower crown, and an equally weak
parastyle. The DP3 and especially the DP4 are
morphologically more like lower crowned molars
than their permanent successors, with strong
styles and PW shorter than AW (Fig. 16). As in the


permanent premolars, no posterolabial cingula are
present on the deciduous premolars.
Lower cheekteeth (Figs. 12B, D, 14C, 17;
Table 6) have the classic bilophodont morphology
of lapirus. Hypolophids on premolars are complete
and in unworn teeth are only slightly lower than
the protolophids (Figs. 12B, 17A). Protolophids
and hypolophids on unworn molars are of equal
height. The p2 has a well developed metalophid
(cristid obliqua) and paralophid; the former blocks
the transverse valley between the protolophid and
hypolophid. Average p2 L/p3 L is 112 percent (N =
12; OR = 104-119). The p3 also has a pronounced
metalophid, but it is not as tall as that of the p2.
L of the p3 is similar to that of the p4 (Table 6),
but the p3 is on average narrower, such that mean
PW/L is 80 percent in p3 and 87 percent in p4 (N
= 14). AW is always less than PW in the p3, with a
mean AW/PW of 91 percent. AW is slightly greater
relative to PW in the p4, with a mean AW/PW of
94 percent, and AW slightly exceeds PW in 2 of 13
individuals (UF 244502 and 224672). The p4-m3
lack a metalophid. Prominent anterior and p,- te,- f--
cingulids are present on unworn and slightly
worn p3-m3, but lateral and lingual cingulids are
absent. The anterior and posterior cingulids are
soon obliterated by interdental wear (Fig. 17). The
posterior cingulid of the m3 lacks a prominent
hypoconulid, and instead has only a slightly
thickened central stylid. With rare exceptions (e.g.,
UF 206876, 206878), AW on ml and m2 exceeds
PW, with very similar mean AW/PW ratios, 104
percent in ml (N = 16) and 103 percent in m2 (N =
12). The PW of the m3 is relatively narrower than
that of the ml and m2, with a mean AW/PW of 111
percent (N= 12).
The lower deciduous premolars are on
average longer, narrower, and less high crowned
than their permanent replacements (Fig. 18; Table
7). The dp2 has a strong metalophid like the p2, but
the dp3 does not. The hypolophids are complete
and as tall as the protolophids on the dp2-dp4.
Small metastylids are present in unworn and
slightly worn dp2-dp4 on the posterolingual side
of the metaconids; these are rapidly lost to wear.
The dp3 and dp4 are molariform in their overall






HULBERT JR.: Blancan Tapirus from Florida


Figure 15. Adult upper cheek teeth of Tapirus lundeliusi sp. nov. in occlusal view. A, UF 224674, left
P1-M3, from Haile 7G; B, UF 224680, leftPl-M3, from Haile 7G; C, UF 115970, leftPl-P4, from Inglis
lA; D, UF 18175, right Ml-M2 (reversed), from Inglis lA; E, UF 97116, right P2 (reversed), from Inglis
1B; F, UF 242909, partial right M3 (reversed) from Withlacoochee River lA.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. f,, )


Table 5. Measurements of individual cheek teeth of the holotype, paratype, and
77,,, hmndeliusi sp. nov. from the Haile 7C locality, Alachua County, Florida.


referred specimens of


160715
160715
160715
160715
160715
160715
160715
121736
121736
121736
121736
121736
121736
121736
206876
206876
206876
206876
206876
206876
206876
206878
206878
206878
206878
206878
206878
206878
212272


Locus L AW


Locus
P1
P2
P3
P4
Ml
M2
M3
PI
P2
P3
DP4
M1
M2
M.
PI
P2
P3
P4
M1
M2
M3
P1
P2
P3
P4
M1
M2
M3
M1


L
18.5
19.1
19.4
20.8
21.9
24.4
25.3
18.1
19.4
19.9
20.9
22.1
23.7
25.6
17.2
19.8
20.2
21.3
22.0
23.7
24.7
19.3
19.3
20.1
21.2
21.6
24.5
25.3
22.4


AW PW
- 17.4
21.3 22.4
23.7 23.0
25.6 24.5
26.3 22.5
28.0 23.3
28.2 22.6
- 16.2
20.6 22.2
24.1 22.7
24.8 21.5
26.0 22.9
25.8 22.2
26.5 22.1
-- 15.2
21.2 22.9
23.8 23.4
24.3 23.8
25.8 23.1
27.0 23.5
26.8 22.1
- 15.8
21.8 24.1
25.3 24.5
26.6 25.5
26.1 22.5
27.6 24.0
27.8 22.3
26.7 23.8


p2
p3
p4
ml
m2
m3


p2
p3
dp4
ml
m2
m3


p2
p3
p4
ml
m2
m3
M-1

p2
p3
p4
ml
m2
m3


21.9
19.1
21.0
21.9
24.1
24.6


23.8
21.0
22.2
22.9
23.7
25.8


23.4
20.8
21.4
22.5
24.0
26.0


22.7
21.5
21.7
23.3
25.4
26.2


14.4
16.8
17.0
18.1
18.3



14.3
15.5
17.5


18.1



15.4
17.1
17.4
18.0
18.3



15.2
17.4
17.5
18.8
18.6


PW


14.9
16.0
18.5
16.8
17.3
17.2


14.4
16.0
15.6
16.8


17.5


14.6
17.0
18.8
17.8
18.2
16.4


15.4
17.2
19.7
18.1
19.2
18.0






HULBERT JR.: Blancan Tapirus from Florida


3 cm
B















3 cm










Figure 16. Juvenile and subadult upper cheek teeth of Tapirus lundeliusi sp. nov. from Haile 7C and 7G in
occlusal view. A, UF 224673, right DPl-DP4; B, UF 224662, right DPl-Ml; C, UF 121736, right Pl-P3,
DP4, M1-M2. Upper scale bar applies to A and B; lower scale bar to C.








Table 6. Univariate statistics for permanent cheekteeth of Iapirus hmdeliusi sp. nov. from the early Pleistocene of Florida and Tapirus
veroensis from the middle and late Pleistocene of Florida, Georgia, and South Carolina. Rightmost two columns show the results of two-
tailed t-tests for significant differences of the means for each tooth measurement. NS = not significant (i.e., p > 0.05).


P1 L
P1 PW
P2 L
P2 AW
P2 PW
P3 L
P3 AW
P3 PW
P4 L
P4 AW
P4 PW
MI L
MI AW
MI PW
M2 L
M2 AW
M2 PW
M3 L
M3 AW
M3 PW


lapirus veroensis


x
18.56
17.00
19.44
21.45
23.51
19.97
24.64
24.08
21.35
26.60
25.33
22.35
26.56
23.41
24.57
28.29
24.33
25.23
27.84
22.86


s
0.86
1.00
0.84
0.90
0.87
0.55
0.77
0.89
0.60
0.87
0.81
0.76
0.56
0.62
0.96
1.02
1.01
0.85
0.96
1.17


', i MAX
17.0 19.8
15.2 18.4
18.0 21.6
19.7 22.9
22.2 25.1
18.6 20.7
23.7 26.1
22.7 25.7
19.9 22.2
24.3 27.5
23.8 26.9
20.7 23.9
25.8 27.7
22.5 24.9
22.7 26.2
25.8 29.6
22.2 26.0
23.4 27.0
26.2 29.2
21.1 25.8


cv
4.65
5.88
4.34
4.21
3.71
2.74
3.14
3.68
2.81
3.26
3.20
3.41
2.12
2.67
3.89
3.60
4.17
3.38
3.45
5.11


x
18.77
16.44
19.84
21.31
23.56
20.49
24.74
24.69
21.65
26.93
26.27
22.78
26.48
24.07
24.93
28.75
25.80
24.80
28.48
24.07


s
1.06
1.53
0.97
1.04
1.23
0.94
1.08
1.16
1.25
1.37
1.40
1.16
1.16
1.02
1.16
1.42
1.17
1.24
1.65
1.44


MIN
15.7
13.7
18.0
18.9
21.4
18.9
22.4
22.3
18.7
24.0
23.0
20.2
24.1
22.2
22.8
25.0
23.1
22.0
24.7
20 3


MAX
20.6
19.4
21.9
23.3
25.8
22.6
26.6
26.7
23.6
29.3
28.9
25.5
28.8
26.3
27.6
31.9
28.3
27.9
31.7
27 0


lapirus ', '. ,, sp. nov.


(CV
5.67
9.31
4.90
4.89
5.21
4.60
4.35
4.70
5.76
5.10
5.33
5.09
4.39
4.24
4.67
4.94
4.53
5.02
5.79
5.98


t
-0.699
1.334
-1.513
0.479
-0.134
-2.166
-0.322
-1.839
-0.963
-0.872
-2.553
-1.456
0.287
-2.520
-1.185
-1.256
-4.516
1.283
-1.422
-2.971


significance
NS
NS
NS
NS
NS
p<0.05
NS
NS
NS
NS
p<0.05
NS
NS
p<0.05
NS
NS
p<0.001
NS
NS
p<0.05









Table 6. Continued.


Iapirus hmndeliusi sp. nov.


m


Iapiru~s i'eroeisis


x
23.55
14.71
21.00
15.33
16.80
21.67
17.65
18.78
23.13
18.16
17.35
25.15
19.27
18.61
26.32
19.23
17.32


s MIN
0.89 21.9
0.94 12.8
1.10 19.1
0.66 14.3
0.57 16.0
1.04 19.8
0.70 16.4
0.68 17.5
0.99 21.5
0.82 17.0
0.67 16.1
1.23 23.4
0.91 18.0
0.91 17.3
1.20 24.6
1.10 18.1
0.70 16.4


p2 L
p2 PW
p3 L
p3 AW
p3 PW
p4 L
p4 AW
p4 PW
mlL
ml AW
ml PW
m2 L
m2 AW
m2 PW
m3 L
m3 AW
m3 AW


MAX
24.7
15.9
23.2
16.4
17.7
23.6
19.3
19.9
25.2
20.0
18.5
26.8
21.7
20.5
29.3
21.9
18.8


N x s
10 23.62 1.12
38 14.88 1.18
17 21.39 1.21
18 15.97 1.18
18 18.02 1.37
18 22.24 1.29
17 18.94 1.27
16 20.27 1.46
50 22.55 1.07
18 18.71 1.01
19 17.56 0.95
56 25.45 1.23
55 20.18 0.97
55 19.21 1.03
53 26.63 1.38
52 19.95 1.00
53 17.90 0.95


MEN
20.9
11.8
18.0
14.1
15.5
19.6
15.2
17.4
20.3
16.6
15.8
21.8
17.8
16.0
23.9
18.0
16.2


CV t significance


3.79
6.40
5.23
4.31
3.38
4.82
3.96
3.64
4.26
4.54
3.84
4.89
4.70
4.88
4.56
5.72
4.02


MAX
25.5
17.3
24.6
19.0
20.9
24.5
21.2
23.4
25.8
20.8
20.6
27.6
22.3
21.8
29.8
22.8
20.3


4.75
7.91
5.65
7.40
7.57
5.78
6.73
7.22
4.77
5.40
5.40
4.83
4.80
5.35
5.19
5.01
5.33


-0.232
-0.474
-1.127
-1.953
-3.259
-1.519
-3.496
-3.678
1.956
-2.011
-0.801
-0.811
-3.113
-1.914
-0.737
-2.247
-2.070


NS
NS
NS
NS
p<0.01
NS
p<0.01
p<0.01
p<0.05
p<0.05
NS
NS
p<0.01
NS
NS
p<0.05
p<0.05






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)


5 cm


Figure 17. Lower cheek teeth of Tapirus lundeliusi sp. nov. from Haile 7C and 7G in occlusal view.
A, UF 121736 with right p2-p3, dp4, ml, and erupting m2; B, UF 221720 with left p2-m2 and erupting
m3 (reversed); C, UF 160715 with right p2-m3; D, UF 207868 with right p2-m3; E, UF 224680 with
right p2-m3.


Ad- -19z
AUL






HULBERT JR.: Blancan Tapirus from Florida


2 cm












Figure 18. Left mandible of very young juvenile of Tapirus lundeliusi sp. nov., UF 224664, from Haile 7G
in A, lateral; B, occlusal; and C, medial views. Teeth present are fully erupted but essentially unworn dp2-
dp3, fully formed but unerupted dp4 in crypt, and forming ml in crypt (partially visible through broken
area just anterior to mandibular foramen in C). Upper scale bar applies to A and C, lower scale bar to B.









Table 7. Univariate statistics for deciduous cheekteeth of 'i,,1 I 1,,. ,',, sp. nov. from the early Pleistocene of Florida and Tapirus
veroensis from the middle and late Pleistocene of Florida, Georgia, and South Carolina. Rightmost two columns show the results of two-
tailed t-tests for significant differences of the means for each tooth measurement. NS = not significant (i.e., p > 0.05).


MIN MAX CV


DP1 L
DPI PW
DP2 L
DP2 AW
DP2 PW
DP3 L
DP3 AW
DP3 PW
DP4 L
DP4 AW
DP4 PW
dp2 L
dp2 PW
dp3 L
dp3 AW
dp3 PW
dp4 L
dp4 AW
dp4 PW


x
18.17
16.91
19.86
18.95
20.11
20.73
21.31
20.97
21.56
24.26
21.76
25.67
14.59
21.62
14.78
14.93
23.11
16.07
16.19


0.82
0.60
0.60
0.60
1.07
0.71
0.77
1.10
0.71
0.44
0.86
0.72
0.64
0.66
0.74
0.79
0.61
0.34
0.62


,,. ..' lundeliusi n. sp.


16.6
16.0
18.4
18.0
18.2
19.4
20.3
19.2
20.8
23.3
20.6
24.5
13.6
20.6
13.8
13.6
22.2
15.5
15.3


19.4
18.0
20.9
19.9
21.4
22.0
22.8
22.2
22.8
24.8
23.1
27.0
15.6
22.7
16.1
16.3
24.3
16.5
17.3


4.54
3.55
3.04
3.17
5.33
3.42
3.61
5.23
3.28
1.81
3.96
2.79
4.36
3.04
5.02
5.28
2.65
2.14
3.82


Tapirus veroensis
N x s MIN
5 17.71 0.85 16.4
5 16.48 1.01 15.4
14 20.00 1.03 18.0
13 18.59 0.98 16.5
13 19.96 0.94 18.6
12 20.84 0.95 19.5
12 21.32 0.74 19.7
10 20.82 0.67 19.3
11 22.43 1.03 20.6
11 25.22 0.88 23.8
11 22.53 1.29 20.2
9 25.92 1.19 24.2
10 15.06 0.80 13.8
8 22.26 1.15 20.8
7 15.66 0.92 14.3
8 15.68 1.11 14.1
19 23.32 1.03 21.8
19 17.36 0.86 16.1
20 16.64 0.69 15.6


MAX
18.5
17.6
21.3
20.2
21.2
22.4
22.3
21.7
23.5
26.9
24.0
27.7
16.0
24.1
16.7
17.0
24.7
19.2
18.0


CV
4.82
6.13
5.15
5.28
4.69
4.56
3.48
3.20
4.58
3.51
5.73
4.59
5.28
5.16
5.87
7.07
4.42
4.94
4.13


t
1.065
1.136
-0.447
0.985
0.393
-0.334
-0.030
0.370
-2.256
-3.097
-1.477
-0.599
-1.441
-1.758
-2.336
-1.918
-0.665
-4.743
-1.852


significance
NS
NS
NS
NS
NS
NS
NS
NS
p<0.05
p<0.01
NS
NS
NS
NS
p<0.05
NS
NS
p<0.001
NS






HULBERT JR.: Blancan Tapirus from Florida


proportions and morphology, although differ from
true molars in that mean AW and PW are about the
same (Figs. 17A, 18B). Isolated dp4s and m1s can
be distinguished on the basis of AW/L ratios; those
with ratios less than 73 percent are dp4s, greater
than 73 percent, mis. Lower crown height and
thinner enamel also characterize the dp4 relative to
the m 1.
DESCRIPTION AND COMPARISONS
OF TAPIRUS SP. FROM SOMMERS PIT,
WALDEN PIT 1, KISSIMMEE RIVER 6,
HUBBRD PIT, AND US 19 BRIDGE SITE
A species of Tapirus smaller in average size
than any extant species is found at five Blancan
localities in Florida. The isolated p4 from Sommers
Pit (UF/TRO 1469) is unworn and probably had
not fully erupted prior to death (Fig. 19D). The
hypolophid is very slightly lower in height than
the protolophid. The tooth is smaller than any p4
referred to 7,'o ir hndeliusi. Its L does fall within
the OR of T lundeliusi, but that is misleading as
the anterior and posterior cingulids have not been
removed by interdental wear with the adjacent
teeth. Its L is only matched by those of T. i,. i,, i./
with well worn teeth, such as UF 224680. The AW
of UF/TRO 1469 (15.1 mm) is about 3.5 standard
deviations less than the mean of T lundeliusi, while
its PW (16.7 mm) is smaller by about 3 standard
deviations. Both widths are below the minimum
values recorded for T. lundeliusi and 7T'iqA,.:
veroensis (Table 6).
The p3 of UF/TRO 1489 (Fig. 19C) from
Walden Pit 1 is from a smaller, older individual
than UF/TRO 1469. Its L, AW, and PW are all
smaller than the minimum values of p3s of T.
lundeliusi and T. veroensis (Table 6), often by a
large margin. UF 51249 is a moderately worn p2,
with the characteristic paralophid and metalophid
of tapirid p2s (Fig. 19B). Its L and PW are smaller
than the minimum values of p2s of T hmdeliusi and
the minimum p2 L for T veroensis (Table 6). The
narrowest p2s of T. veroensis have widths similar
to UF 51249, but such specimens are rare. The PW
of UF 248788 is identical to that of UF 51249; L
can not be measured due to breakage.


UF 249089 (Fig. 19E-F) from Hubbard
Pit is a slightly worn dp3 (or, less likely, a dp4)
that is missing the roots; minor damage to the
anterolingual base of the crown prevents accurate
determination of its AW. Both the L and PW of UF
249089 fall below the OR of either the dp3 or the
dp4 of T. lundeliusi and T. veroensis (Table 7). The
ratio of PW/L in UF 249089 is 0.67. The dp3s of
T lundeliusi have similar ratios (x = 0.685, N =
15, OR = 0.64-0.74), while the dp4s are slightly
broader (x = 0.701, N = 12, OR = 0.68-0.73). Most
(22 of 27) dp3s and dp4s of T veroensis are also
slightly broader, with PW/L > 0.70.
The navicular (UF 247180) has the
characteristic shape of this element in Iapirus, but
is of small size. Greatest transverse width of its
astragalar facet, 29.8 mm, is 4.0 standard deviations
below the mean for a sample of naviculars of T
!,,it3, her.. and well below its observed range
(x = 35.7; s = 1.49; OR = 31.9-38.1; N = 22).
Slight breakage on UF 247180 prevents accurate
measurement of the depth of the astragalar facet,
but the value would have been very close to 21
mm. Again, this value lies outside the range for T
!,,.A, h... (x= 25.5; s = 1.25; OR= 23.5-28.7; N=
22). In contrast, the dimensions of UF 247180 are
very similar to those of the Gray Fossil Site sample
of Tapirus polkensis (astragalar facet width: x =
29.2; s = 1.96; OR = 26.1-33.3; N= 17; astragalar
facet depth: x = 21.3; s = 1.14; OR = 20.0-23.9; N
=17).
Small tapirs, those significantly smaller in
body size than any extant species, are known from
the Hemphillian (late Miocene-early Pliocene) of
Florida and Tennessee (Fig. 19A; Hulbert 2005;
Hulbert et al. 2009), but have not previously been
reported from the Pli;r,- c of North America.
The p3 of UF/TRO 1489 is larger than average
members of the Tennessee sample of lapirus
polkensis, but falls within that population's OR
(Hulbert et al. 2009). When compared to the two
species of Twq ii -,. in the late Hemphillian Palmetto
Fauna of Florida (Webb et al. 2008), the L and AW
of UF/TRO 1489 more closely resemble those of
the larger, unnamed species, while its PW is more
similar to that of T. ,/ -/,... In contrast, LUF/TRO






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)


1469 is larger than p4s referred to T.polkensis from
both Florida and Tennessee (Hulbert et al. 2009),
and its L, AW, and PW are all within the ORs of the
larger, unnamed Palmetto Fauna tapir. The L of the
p2, UF 51249, is greater than any referred specimen
of T polkensis from Florida or Tennessee (n = 31),
and instead is very similar to the two known p2s of


the larger, unnamed species in the Palmetto Fauna
(19.3 for UF 23948 and 20.1 for UF 124191). Its
PW (along with that of UF 248788), however, is
slightly less than those of these two specimens,
and within the range of T polkensis. The L and
PW of UF 249089 and L of UF 255358 exceed
those of dp3s of T polkensis from both Florida


1 cm


Figure 19. Lower premolars of Tapirus sp. indet. from Florida in occlusal (A-E) and lingual (F) views. A,
UF 124191, left partial mandible with p2-p4 from Whidden Creek Site, Fort Meade Mine, Polk County,
Florida, very late Hemphillian; B, UF 51249, left p2 from Kissimmee River 6; C, UF/TRO 1489 right p3
(reversed) from Walden Pit 1; D, UF/TRO 1469 right p4 (reversed) from Sommers Pit; E, occlusal and
F, lingual, views of UF 249089, left dp3 from Hubbard Pit. The Blancan teeth in B-F are too small to be
referred to Tapirus lundeliusi or Tapirus haysii, but are instead within the size range of an undescribed,
small species from the late Hemphillian of Florida. Upper scale bar applies to A-D; lower scale bar to E-F.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. f,, )


In a cladistic analysis using all of the taxa
in Appendix 3 with the L .,_,.*t, of Tapirus
hezhengensis, ordered and unordered multi-state
characters produced the same relationships among
in-group taxa (Fig. 20). With unordered characters,
there are two most parsimonious cladograms,
each with an overall consistency index of 0.68
and retention index of 0.74. The analyses strongly
support a monophyletic clade comprised of T.
hlundeliusi, Tapirus haysii, and Tw,'u ir. veroensis
(i.e., the subgenus iJ ,,i;_ ~:), and that T.
hlundeliusi is the sister group to T. haysii plus T.
veroensis. The interrelationships of lapirus bairdii
and Tapirus j ,1.. / .Vi* with Helicotapirus are
unresolved, but there is strong support that they
are more closely related to H.... l/.., iJs. than
to other species of lapirus. Synapomorphies of
T. (Helicotapirus) are concave lateral surface of
lacrimal (character 30 in Appendix 2); anterior
projection of ascending ramus of mandible (39);
increased depth of horizontal ramus below molars
(40); enlarged parastyles on upper molars (55);
loss of posterolabial cingula on upper cheekteeth
(56); and increased relative length of the p2 (69).
Synapomorphies of T. haysii and T. veroensis
relative to T. hlundeliusi are delayed fusion of
interparietal with occipital (9); relatively short
nasal bone (10); absence of a maxillary flange (26);
increased length of lacrimal (29); taller metaloph
on P2 (51); and reduced p3 cristid obliqua
(73). The clade consisting of Helicotapirus, IT.
h?..i;J,, and T. polkensis is supported by seven
unambiguous synapomorphies: broad dorsal table
of frontal (4, reversed in T. hmdeliusi); triangular
interparietal (8); meatal fossa on nasal deep (14)
and approaches midline of skull (15); meatal fossa
on frontal extensive with a distinct posterior margin
(16, reversed in T. limnd.J. ..), broad and flat
posterior lacrimal process (33); and relatively short
postcanine diastema (37, reversed in T bairdii). The
trichotomy among Helicotapirus, T. bairdii, and IT.
" '/.. ur~"-. is not resolved in this analysis. To date no
North American fossil species of Iapirus has been
described that shares some of the autapomorphic
character states of T. bairdii such as presence of
a broad sagittal table instead of a narrow sagittal


crest, nasal notch extending posterior to orbit,
loss of descending sigmoid process of nasal bone,
greatly expanded maxillary flange, ossification
of nasal septum, shortened posterior process of
premaxilla, very long postcanine diastema, and
small cheektooth dimensions relative to body size
(and skull length). The distribution of cranial and
dental characters of such a fossil .....- could
resolve the trichotomy.
Addition of Tapirus hezhengensis to the
analysis does not effect support for the following
monophyletic clades: Tapirus lundeliusi, Tapirus
haysii, and lapirus veroensis (and their relative
positions); Tapirus (1..-. P. ,,. Tapirus bairdii
and Iapirus ,!i-, I and lapirus terrestris plus
Tapirus j'i._h..i.....: (Fig. 21). Two differences
are found from the first analysis. The trichotomy
between T (H.-. ',,i' ,), T bairdii and T
polkensis is resolved, albeit weakly (Bremer support
of 1), with T bairdii sharing a closer relationship
with T. (Helicotapirus) than does T polkensis.
Second, the possibility of a close relationship
between lapirus webbi from the late Miocene
Florida with the two extant South American species,
a hypothesis first tentatively proposed by Hulbert
(1995; then using the name Tapirus simpsoni for
the samples later named T. webbi), is revived (Fig.
21A). A late Miocene dispersal of T. webbi or a
closely related form into South America, followed
by in situ evolution that produced T terrestris and
1:. / iAn ;.h,, is congruent with molecular evidence
of a relative ancient (pre-Pliocene) divergence
time between 7' bairdii and the clade formed by
T terrestris and T. //; .- huqi: (Ashley et al. 1996;
Norman & Ashley 2000). More recently, Rinc6n
et al. (2009) noted the resemblance between some
fragmentary tapirid material from the late Pliocene
or early Pleistocene of Venezuela and T. webbi.
Although their geologic and chronologic evidence
remains controversial, Campbell and co-authors
have persistently proposed a late Miocene dispersal
to South America of a few North American
ungulates, including a tapir (Campbell et al. 2000,
2006, 2010), coeval with the well accepted first
d,i,- p!, -,-I of ground sloths from South America to
North America (e.g., Morgan 2005). A much later






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


degree of intraspecific variation in a number of
features that show greater consistency in extant
species and Japirus veroensis, for example in the
form of the sagittal crest and number of lacrimal
foramina (Hulbert et al. 2009). Such is not the case
with the Haile 7G sample of Tapirus hmlundeliusi,
which is almost as numerous as the Gray Site
sample; once ontogenetic variation is factored,
its intraspecific variation in cranial morphology
is similar to that observed in extant species of
7,',//r Characters that present a high degree of
variation in the Haile 7G population, e.g. shape of
the nasal bone and development of the lingual cusp
and transverse loph on the PI, are also variable in
large samples of all species of 7T'q i,, However,
many skulls and mandibles from Haile 7G remain
to be prepared, so this observation must remain
tentative for now.
One of the unusual aspects of the Haile 7C
and 7G sample of Tapirus lundeliusi is the rarity of
old adult individuals. Only one out of 44 individuals
that can be aged is in this age class (UF 244600),
and it only has very small areas of exposed dentine
on its m3, so it barely falls into this category. Tapirs
from the old adult wear stage are not uncommon
at other Florida fossil localities (pers. obs.) or at
the Gray Fossil Site (Hulbert et al. 2009). Indeed,
the lone tooth from the Withlacoochee River 1A
sample of T. lundeliusi is at the old adult stage as
is UF/TRO 1476 from Walden Pit 2. It is as yet
unclear whether the taphonomic factors causing
the very large sample of tapir skeletons at the
two Haile 7 sites were acting independent of age
("catastrophically", although not likely a single,
rapid event), or were actively biased against older,
more experienced individuals. This will probably
not be resolved until the cause of the mortality of
the Haile tapirs is better understood.
Although there is broad overlap in all tooth
dimensions, Tapirus lundeliusi is significantly
different from Tapirus veroensis in 13 out of 37
tooth measurements in the permanent dentition
(Table 6) and four out of 19 in the deciduous
dentition (Table 7). One consistent dissimilarity is
that posterior widths of P4-M3 are all significantly
smaller in T. hmlundeliusi. The other difference is the


significantly smaller widths of the p3 (PW only)
and p4 (AW and PW) in T. hmlndeliusi. That the ml
of T. hlundeliusi is significantly longer than that of T
veroensis is likely an artifact of the demographics
of the two measured samples. The sample of T.
lundeliusi includes a much greater proportion of
juveniles and subadults with unworn and slightly
worn mls which have not lost their anterior and
posterior cingulids, while the T. veroensis sample
contains a greater proportion of full and old adults in
which the ml has worn against the adjacent p4 and
m2, reducing its measurable length. Including only
unworn or slightly worn m 1s from the T. veroensis
sample produces a population with a mean L ..f __' 9
mm (N = 12) that is not significantly different from
that of T limJ. iu'-./. The greatest differences in the
deciduous dentition are the significant smaller AW
of DP4 and dp4 in T7 lundeliusi (Table 7).
In their discussion of the type locality of
Tapirus haysii, Ray and Sanders (1984) noted
that the USNM collection contained specimens
of both T. haysii and Tapirus veroensis from the
Neuse River near New Bern, North Carolina.
The associated fauna was only briefly noted
(Ray & Sanders 1984:293), but resembles mixed
Blancan-Rancholabrean faunas in Florida, such
as Santa Fe River 1 (compare with MacFadden &
Hulbert 2009:table 1). For example the presence
of ,; ii:i"'-j',," and T7 haysii are evidence of a
Blancan component, while !..1Aiunim., Bison,
and Castoroides together indicate a Rancholabrean
age. Thus the specimens Ray and Sanders (1984)
identified as T. veroensis may simply be standard
Rancholabrean specimens of that taxon. However,
given that there is a now a second species of Tapirus,
of similar size as T. veroensis, from the Blancan,
then the smaller Neuse River tapir specimens need
to be critically compared with T /i ii.. lundeliusi.
Although currently known from a limited
geographic range in Florida, it is likely that future
discoveries of late Blancan faunas elsewhere in the
southeastern United States will produce specimens
of hapirus lundeliusi. Fossil species of Tapirus tend
to have rather broad geographic distributions, as do
three ofthe four extant species. Even ifits geographic
range is currently small, it has broad implications






HULBERT JR.: Blancan Tapirus from Florida


Figure 22. Mounted skeleton of Tapirus lundeliusi sp. nov. on public display at the Florida Museum of
Natural History. Postcranial skeleton belongs to UF 206878 (with minor restoration). Skull and mandibles
(excluding symphysis and incisors) are casts of UF 160715 (with restoration); mandibular symphysis and
incisors are a cast of UF 206878 (see Fig. 14A). Skeletal pose designed by Gina Gould; skeleton mounted
and casts made and painted by Steve and Sue Hutchens; photography by Jeff Gage.


for the phylogeny of Tapirus. The mixture of
plesiomorphic character states in T lundeliusi (e.g.,
presence of maxillary flange, narrow lacrimal, long
nasal, early fusion of interparietal) together with
apomorphies shared with Tapirus veroensis and
Tapirus haysii (e.g. short postcanine diastema,


deep horizontal ramus, absence of posterolabial
cingula on upper cheekteeth) allows a better sense
of the evolutionary relationships and character
state transformations between Miocene species
such as Tapirusjohnsoni and Tapiruspolkensis and
the later Pleistocene species.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. t,, )


CONCLUSIONS
T..,p,. are a common component of the large
mammalian herbivore guild in Florida during
the late Blancan (early Pleistocene). Two named
-.....i.--. of TKq ii-. are recognized from this
interval, the very large Japirus haysii and the
medium-sized T_7i/,,,. lundeliusi sp. nov. (Fig.
22), plus a smaller, unnamed form making a
total of three. But there is only possible one case
when more than one species has been found at the
same locality (Kissimmee River 6 with T haysii
and the small, unnamed species), -.-*---1._,.
some type of ecologic .f.:,i- ;-.,n or ,-i-, F e;;.. -
exclusion was the norm. As the Kissimmee River
specimens were collected from spoil piles created
by dcI.~,i it is not certain that the two taxa are
in fact contemporaneous at this locality. Tapirus
hmdeliusi is the best Iill<- c'.1 late Blancan
tapir from Florida, recognized from eleven sites,
although complete -"-"-iII ti t- derive from just two,
Haile 7C (the type locality) and Haile 7G. 'ql ,, -..
haysii is rarer and known from only a few Blancan
sites in Florida (Fig. 1; Appendix 1). T haysii
became more common and widespread in Florida
during the early Irvingtonian (Hulbert 1995). In
contrast to 7. haysii, no Irvingtonian records of T'
lundeliusi are known, with the species apparently
becoming extinct near the end of the Blancan. The
third late Blancan species, known from Walden Pit
1, Sommers Pit, Hubbard Pit, Kissimmee River 6,
and the US 19 Bridge site, is significantly smaller
than I hlundeliusi, and may represent a holdover of
an undescribed species from the late Hemphillian
of Florida.
Almost all standard measurements on the
skull, mandible, and teeth show considerable
c',,'-il p between ]apirus hmndeliusi and the
common middle to late Pleistocene species T7i ,i !
veroensis. But T. h,,.L 'i;,.. retains a number of
plesiomorphic features in its skull that are more
derived in 7' veroensis (and Tapirus haysii), such
as having a maxillary flange and a smaller lacrimal.
Formation of the sagittal crest and fusion of the
interparietal bone with the occipital and parietals
occur earlier in ontogeny in T lundeliusi compared
to TI veroensis (using tooth _.,l'.' sequences to


age specimens). I. Ih ; hi.i shares with :. haysii
a much greater relative mandibular condyle height
than found in 7. veroensis or other species of
Tapirus.
Phylogenetic analysis using 39 characters
provides strong support for a North American clade
composed of Tapirus lundelusi, lapirus veroensis,
and Tapirus haysii. T. lundeliusi is the sister taxon
to the other two. This clade is formally recognized
as a new subgenus, H.. .i,q'ii .. Phylogenetic
analysis also strongly supports a clade comprised
of Helicotapirus, the late Miocene-early Pliocene
species lapirus polkensis, and the extant species
Tapirus bairdii. Many of the synapomorphies of
this clade are osteological features of the nasal
and frontal bones related to the development of
the meatal diverticulum, a soft tissue structure
of uncertain function (Witmer et al. 1999). The
H. i. ,..qv,;i :. 77 bairdii-T7 polkensis clade most
likely diverged from that containing the other two
extant New World species (Japirus terrestris and T.
/ in/.i in.._ ) in the late Miocene.

ACKNOWLEDGMENTS

This is University of Florida Contribution to
Paleobiology number 624. Any project spanning
two decades has received assistance and
encouragement in many ways. Early research
and travel to collections to obtain comparative
measurements were undertaken during 1994-1999
while the author was on the faculty of Georgia
Southern University and funded by the Georgia
Southern University Foundation and Faculty
Research Subcommittee. Excavations at Haile
7C in 1994-1995 were funded by NSF grant
EAR-940 '_- to S. D. Emslie and those at Haile
7G in 2006-2008 were funded by NSF grant
EAR-0639307 to J. Bloch and RCH. Permission
to collect fossils at Haile 7C and 7G was provided
by L. Rogers and Limestone Products, Inc. Many
hundreds of public volunteers worked with dozens
of UF personnel and students at Haile 7C and 7G
to collect a vast number of fossil skeletons. Special
praise goes to G. Morgan and A. Poyer for their
extensive work with RCH at Haile 7C in 1989-
1990; the other members of the 1994-1995 field






HULBERT JR.: Blancan Tapirus from Florida


crew at Haile 7C (S. Emslie, A. Pratt, B. and R.
Toomey, and S. and S. Hutchens); R. Portell for his
discovery of the Haile 7G locality; and my primary
field associates at Haile 7G: A. Poyer, J. Bloch, J.
Bourque, A. Hastings, and E. Simons. J. Waldrop
discovered and donated specimens from Sommers
and Walden Pits. S. and S. Hutchens discovered
and donated specimens from the Inglis ID and
Waccasassa River 9A sites. A. Gipson discovered
and donated specimens from the Withlacoochee
River 1A site. The Kissimmee River, US 19 Bridge
Site, and Hubbard Pit specimens were donated by
H. Converse, A. Kerner, and J. Smith, respectively.
Other recent donors of tapir fossils to UF include
S. Alter, M. Anderson, J. Bryan, R. Dykeman, L.
Geary, C. Jeremiah, H. Means, M. Renz, P. Roth,
M. and S. Searle, D. Thulman, B. and J. Toomey,
and the family of the late R. Piotrowski. I thank M.
Colbert, C. Ray, E. Scott, B. Shockey, and S. Wallace
for useful discussions over the years on tapirid
morphology and systematics. Helpful comments
and suggestions for improvement were provided
by J. Miller who served as editor of this issue and
by two outside reviewers. And I especially wish to
thank E. Lundelius for his guidance and mentoring
during the beginning phase of my career.
LITERATURE CITED
Ashley, M. V., J. E. Norman, & L. Stross. 1996.
Phylogenetic analysis of the perissodactylan
family Tapiridae using mitochondrial cyto-
chrome c oxidase (COII) sequences. Journal
of Mammalian Evolution 3(4):315-326.
Bader, R. S. 1957. Two Pleistocene mammalian
faunas from Alachua County, Florida. Bulletin
of the Florida State Museum, Biological
Sciences 2(5):53-75.
Bell, C. J., E. L. Lundelius Jr., A. D. Barnosky, R.
W. Graham, E. H. Lindsay Jr., D. R. Ruez,
Jr., H. A. Semken Jr., S. D. Webb, & R. K.
Zakrzewski. "(n I The Blancan, Irvingtonian,
and Rancholabrean mammal ages. Pp. 232-
314 in M. 0. Woodburne, ed. Late Cretaceous
and Cenozoic Mammals of North America.
Columbia University Press, New York.
Boeuf, 0. 1991. Tapirus,/...,i/ ,.ii.. ufm nov. sp.,


nouvelle espece de Tapiridae (Perissodactyla)
du Pliocene de Charente (France). Palaeon-
tographica, Abteilung A 217:177-194.
Brisson, A. D. 1762. Regnum Animale in Classes
IX distributum sive synopsis methodica. Edito
altero auctior. Theodorum Haak, Leiden,
Netherlands, 294 pp.
Campbell Jr., K. E., C. D. Frailey, & L. Romero-
Pittman. 2000. The late Miocene gomphothere
Amahuacatherium peruvium (Proboscidea:
Gomphotheriidae) from Amazonian Peru:
implications for the Great American Faunal
Interchange. Institute Geologico Minero y
Metalutrgico, Serie D: Estudios Regionales,
Boletin 23:1-152.
Campbell Jr., K. E., C. D. Frailey, & L. Romero-
Pittman. 2006. The Pan-Amazonian Ucayali
Peneplain, late Neogene sedimentation in
Amazonia, and the birth of the modern
Amazon River system. Palaeogeography,
Palaeoclimatology, Palaeoecology 239:166-
219.
Campbell Jr., K. E., D. R. Prothero, L. Romero-
Pittman, F. Hertel, & N. Rivera. 2010.
Amazonian magnetostratigraphy: dating
the first pulse of the Great American Faunal
Interchange. Journal of South American Earth
Sciences 29:619-626.
Carpenter, W. M. 1846. Remarks on some fossil
bones brought to New Orleans from Tennessee
and Texas. American Journal of Science,
second series 1:244-250.
Colbert, M. W. 2005. The facial skeleton of the
early Oligocene Colodon (Perissodactyla,
Tapiroidea). Paleontologica Electronica
8:12A. Retrieved from http://palaeo-
electronica.org/2005_ 1/colbertl2/issuel_05.
html
Cope, E. D. 1871. Preliminary report on the
Vertebrata discovered in the Port Kennedy
bone cave. Proceedings of the American
Philosophical Society 12:73-102.
Dalquest, W. W. & Schultz, G. E. 1992. Ice Age
Mammals of Northwestern Texas. A\ l[i.. i-.. .1
State University Press, Wichita Falls, Texas,
309 p.






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


Deng, T., W. He, & S. Q. Chen. 2008. A new
species of late Miocene tapir (Perissodactyla,
Tapiridae) from the Linxia Basin in Gansu,
China. Vertebrata PalAsiatica 46(3): 190-209.
[in Chinese with English summary]
DeSantis, L. G. & MacFadden, B. J. 2007.
Identifying forested environments in Deep
Time using fossil tapirs: evidence from
evolutionary morphology and stable isotopes.
Courier Forschung-Institut Senkenberg
258:147-157.
DeSantis, L. R. G. & MacFadden, B. J. 2008.
Stable isotopes and REEs of ungulates and
xenarthrans from Haile 7G: elucidating the
paleoecology of a late Pliocene fossil site from
Florida. Journal of Vertebrate Paleontology
28(3, supplement):71A.
Desmarest, A. G. 1819. Tapirs fossils. Pp. 458-461.
In Nouveau Dictionnaire d'Histoire Naturelle,
Nouvelle Edition. Chez Deterville, Paris, vol.
32.
Emslie, S. D. 1992. Two new late Blancan avifaunas
from Florida and extinctions of wetland birds
in the Plio-Pleistocene. Natural History
Museum of Los Angeles County Science
Series 36:249-269.
Emslie, S. D. 1998. Avian community, climate,
and sea-level changes in the Plio-Pleistocene
of the Florida Peninsula. Ornithological
Monographs, No. 50, 113 p.
Ferrero, B. S., & J. I. Noriega. 2007. A new upper
Pleistocene tapir from Argentina: remarks
on the phylogenetics and diversification of
Neotropical Tapiridae. Journal of Vertebrate
Paleontology 27(2):504-511.
Gibbard, P. L., M. J. Head, M. J. C. Walker, and the
Subcommission on Quaternary Stratigraphy.
2010. Formal ratification of the Quaternary
System/Period and the Pleistocene Series/
Epoch ,i.lk a base at 2.58 Ma. Journal of
Quaternary Science 25(2):96-102.
Gill, T. N. 1865. [Description of the genus
17., ,, a:, iin ]. Proceedings of the Academy
of Natural Sciences Philadelphia 17:183.
Graham, R. W. 2003. Pleistocene tapir from Hill
Top Cave, Trigg County, Kentucky and a


review of Plio-Pleistocene tapirs of North
America and their paleoecology. Pp. 87-118
in B. Schubert, J. I. Mead, & R. W. Graham,
eds. Vertebrate Paleontology of Caves.
Indiana University Press, Bloomington.
Gray, J. E. 1821. On the natural arrangements
of vertebrose animals. London Medical
Repository Review, 15:296-310.
Hershkovitz, P. 1954. Mammals of northern
Colombia, preliminary report no. 7: tapirs
.,!,- 7,',,,..), with a systematic review of
American species. Proceedings of the United
States National Museum 103:465-496.
Hulbert Jr., R. C. 1988. A new Cormohipparion
(Mammalia, Equidae) from the Pliocene
(latest Hemphillian and Blancan) of Florida.
Journal of Vertebrate Paleontology 7:451-
468.
Hulbert Jr., R. C. 1995. The giant tapir, l pirus
haysii, from Leisey Shell Pit 1A and other
Florida Irvingtonian localities. Bulletin of the
Florida Museum of Natural History 37:515-
551.
Hulbert Jr., R. C. 1999. Nine million years of
Iapirus (Mammalia, Perissodactyla) from
Florida. Journal of Vertebrate Paleontology
19(3, supplement): 53A.
Hulbert Jr., R. C. 2005. Late Miocene Tapirus
(Mammalia, Perissodactyla) from Florida,
with description of a new *.l.... -., Tapirus
webbi. Bulletin of the Florida Museum of
Natural History 45(4):465-494.
Hulbert Jr., R. C., J. I. Bloch, & A. R. Poyer. 2006.
Exceptional preservation of vertebrates from
Haile 7G, a new late Pliocene site from
Florida. Journal of Vertebrate Paleontology
26(3, supplement):78A-79A.
Hulbert Jr., R. C., & G. S. Morgan. 1993.
Quantitative and qualitative evolution in
the giant armadillo Holmesina (Edentata:
Pampatheriidae) in Florida. Pp. 134-177
in R. A. Martin & A. D. Bamosky, eds.
Morphologic Change in Quaternary Mammals
of North America. Cambridge University
Press, Cambridge.
Hulbert Jr., R. C., & G. S. Morgan. 2009. Two






HULBERT JR.: Blancan Tapirus from Florida


new late Blancan vertebrate sites from
Florida, with a reassessment of late Pliocene
vertebrate biochronology in Florida. Abstract
Program, Second Annual Meeting of the
Southeastern Association of Vertebrate
Paleontology, Virginia Museum of Natural
History, Martinsville. Retrieved from http://
web. me. com/dooleyclan/SEAVP/SEAVP_
files/Abstract%20book.pdf
Hulbert Jr., R. C. & S. C. Wallace. 2005.
Phylogenetic analysis of late Cenozoic
Tapirus (Mammalia, Perissodactyla). Journal
of Vertebrate Paleontology, 25(3, supple-
ment):72A.
Hulbert Jr., R. C., S. C. Wallace, W. E. Klippel, & P.
W. Parmalee. 2009. Cranial morphology and
systematics of an extraordinary sample of the
late Neogene dwarf tapir, Tapirus polkensis
(Olsen). Journal of Paleontology 83(2):238-
262.
Jefferson, G. T. 1989. Late Cenozoic tapirs
(Mammalia: Perissodactyla) of western North
America. Natural History Museum of Los
Angeles County Contributions in Science
406:1-21.
Jones, D. S., B. J. MacFadden, S. D. Webb, P.
A. Mueller, D. A. Hodell, & T. M. Cronin.
1991. Integrated geochronology of a classic
Pliocene fossil site in Florida: Linking marine
and terrestrial biochronologies. The Journal
of Geology 99(5):637-648.
Kurten, B., & Anderson, E. 1980. Pleistocene
Mammals of North America. Columbia
University Press, New York, 442 p.
Leidy, J. 1859. Descriptions of vertebrate fossils.
Pp. 99-122 inF. S. Holmes, ed. Post-Pleiocene
Fossils of South Carolina. Russell and Jones,
Charleston.
Leidy, J. 1889. Description of vertebrate remains
from Peace Creek, Florida. Transactions
of the Wagner Free Institute of Science of
Philadelphia 2:19-31.
Louys, W., D. Curnoe, & H. W. Tong. 2007.
Characteristics of Pleistocene mega-
fauna extinctions in Southeast Asia.
Palaeogeography, Palaeoclimatology, Pal-


aeoecology 243:152-173.
Lundelius, E. L. 1972. Fossil vertebrates from the
late Phi; w c,- Ingleside Fauna, San Patricio
County, Texas. Report of Investigation,
Bureau of Economic Geology, University of
Texas, No. 77, 74 pp.
Lundelius, E. L., & B. H. Slaughter. 1976. Notes
on American Pleistocene tapirs. Pp. 226-
243 in C. S. Churcher, ed. Athlon: Essays
in Paleobiology in Honour of Loris Shano
Russell. Royal Ontario Museum, Toronto.
MacFadden, B. J., & R. C. Hulbert. 2009.
Calibration of mammoth (\.,,m ,,r )
dispersal into North America using rare earth
elements of Plio-Pleistocene mammals from
Florida. Quaternary Research 71(1):41-48.
Matthew, W. D., & W. Granger. 1923. New fossil
mammals from the Pliocene of Szechuan,
China. Bulletin of the American Museum of
Natural History 48:563-598.
McKenna, M. C., & S. K. Bell. 1997. Classification
of Mammals above the Species Level.
Columbia University Press, New York, 631 p.
Merriam, J. C. 1913. Tapir remains from late
Cenozoic beds of the Pacific coast region.
University of California Publications, Bulletin
of the Department of Geology 7:169-175.
Meylan, P. A. 1982. The squamate reptiles of the
Inglis 1A fauna (Irvingtonian: Citrus County,
Florida). Bulletin of the Florida State Museum
27(1):1-85.
Morgan, G. S. 2005. The Great American Biotic
Interchange in Florida. Bulletin of the Florida
Museum of Natural History 45(4):271-311.
Morgan, G. S. 2008. Vertebrate fauna and
geochronology of the Great American Biotic
Interchange in North America. Pp. 93-140 in
S. G. Lucas, G. S. Morgan, J. A. Spielmann,
& D. R. Prothero, eds. Neogene Mammals.
New Mexico Museum of Natural History &
Science Bulletin 44.
Morgan, G. S., & S. D. Emslie. 2010. Tropical
and western influences in vertebrate faunas
from the Pliocene and Pleistocene of Florida.
Quaternary International 217:143-158.
Morgan, G. S., & R. C. Hulbert Jr. 1995. Overview






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


of the geology and vertebrate biochronology of
the Leisey Shell Pit local fauna, Hillsborough
County, Florida. Bulletin of the Florida
Museum of Natural History 37:1-92.
Morgan, G. S., & J. A. White. 1995. Small
mammals (Insectivora, Lagomorpha, and
Rodentia) from the early Pleistocene (early
Irvingtonian) Leisey Shell Pit local fauna,
Hillsborough County, Florida. Bulletin of the
Florida Museum of Natural History 37:397-
462.
Norman, J. E., & M. V. Ashley. 2000. Phylogenetics
of Perissodactyla and tests of the molecular
clock. Journal of Molecular Evolution
50(1):11-21.
Olsen, S. J. 1960. Age and faunal relationships of
lhpiravus remains from Florida. Journal of
Paleontology 34:164-167.
Owen, R. 1848. Description of teeth and portions
of jaws of two extinct anthracotheroid
quadrupeds (H'-r.p-i.mn; vectianus and Hyop.
bovinus) discovered by the Marchioness of
Hastings in the Eocene deposits on the N.W.
coast of the Isle of Wight: with an attempt to
develop Cuvier's idea of the classification
of pachyderms by the number of their toes.
Quarterly Journal of the Geological Society
of London 4:103-141.
Petuch, E. J. 1982. Notes on the molluscan
paleoecology of the Pinecrest Beds at
Sarasota, Florida, with the description of
P) I,/ .1., a stratigraphically important
new genus (Gastropoda, Melongenidae).
Proceedings of the Academy of Natural
Sciences, Philadelphia 134:12-30.
Ray, C. E. 1964. Tapirus copei in the Pleistocene
of Florida. Quarterly Journal of the Florida
Academy of Sciences 27:59-66.
Ray, C. E. & A. E. Sanders. 1984. Ple.-r,-,-e
tapirs in the eastern United States. Pp. 283-
315 in H. H. Genoways & M. R. Dawson,
eds. Contributions in Quaternary Vertebrate
Paleontology: a Volume in Memorial to John
E. Guilday. Carnegie Museum of Natural
History, Pittsburgh, Special Publication 8.
Ray, C. E., E. Anderson, & S. D. Webb. 1981. The


Blancan carnivore Ni I; ;i, (Mammalia:
Mustelidae) in the eastern United States.
Brimleyana 5:1-36.
Rinc6n, A. D., G. E. Parra, F. J. Prevosti, M. T.
Alberdi, and C. J. Bell. 2009. A preliminary
assessment of the mammalian fauna from
the Pliocene-Pleistocene el Breal de Orocual
locality, Monagas State, Venezuela. Pp. 593-
620 in L. B. Albright III, ed. Papers on Geology,
Vertebrate Paleontology, and Biostratigraphy
in Honor of Michael 0. Woodburne. Museum
of Northern Arizona Bulletin 65.
Robertson, J. S. 1976. Latest Pliocene mammals
from Haile XV A, Alachua County, Florida.
Bulletin of the Florida State Museum 20:111-
186.
Roulin, X. 1829. Memoir pour servir a l'histoire
du tapir; et description dune espece nouvelle
appartenant aux hautes regions de la Cordillere
des Andes. Annales des Sciences Naturelle
Zoologie Paris 17:26-55.
Ruez, D. R. 2001. Early Irvingtonian (latest
Pliocene) rodents from Inglis IC, Citrus
County, Florida. Journal of Vertebrate
Paleontology 21(1):153-171.
Rustioni, M. 1992. On Pliocene tapirs from
France and Italy. Bollettino della Societa
Paleontologica Italiana 31(3):269-294.
Sanders, A. E. 2002. Additions to the Pleistocene
mammal faunas of South Carolina, North
Carolina, and Georgia. Transactions of the
American Philosophical Society 92(5): 1-152.
Schultz, C. B., L. D. Martin, & R. G. Corner.
1975. Middle and late Cenozoic tapirs from
Nebraska. Bulletin of the Nebraska State
Museum 10:1-21.
Sellards, E. H. 1918. The skull of a Pleistocene
tapir including description of a new species
and a note on the associated fauna and flora.
Annual Report, Florida Geological Survey
10:57-70.
Shunk, A. J., S. Driese, J. Farlow, R. Hulbert, & M.
Whitelaw. 2006. High-resolution stratigraphy
and sedimentology of late Miocene-
Pliocene paleolacustrine strata deposited
in paleosinkhole settings, Eastern U.S.






HULBERT JR.: Blancan Tapirus from Florida


Abstracts with Programs, Geological Society
of America 38(7):85.
Simpson, G. G. 1929. Pleistocene mammalian
fauna of the Seminole Field, Pinellas County,
Florida. Bulletin of the American Museum of
Natural History 56:561-599.
Simpson, G. G. 1945. Notes on Pleistocene and
Recent tapirs. Bulletin of the American
Museum of Natural History 86(2):33-82.
Smith, K. S., & R. L. Cifelli. 2000. A synopsis
of the Pleistocene vertebrates of Oklahoma.
Oklahoma Geological Survey Bulletin, No.
147, 36 pp.
Tedford, R. T., X.-M. W.A,. & B. E. Taylor.
2009. Phylogenetic systematics of the
North American fossil Caninae (Carnivora:
Canidae). Bulletin of the American Museum
of Natural History 325:1-218.
Tong, H. W. 2005. Dental characters of Quaternary
tapirs in China, their significance in
classification and phylogenetic assessment.
Geobios 38(1):139-150.
Tong, H. W., J. Y Liu, & L. G. Han. 2002. On
fossil remains of early Pleistocene tapir
(Perissodactyla, Mammalia) from Fanchang,
Anhui. Chinese Science Bulletin 47(7):586-
590.
Webb, S. D. 1974. Chronology of Florida
Pleistocene mammals. Pp. 5-31 in S. D.
Webb, ed. Pleistocene Mammals of Florida.
University of Florida Press, Gainesville.
Webb, S. D., R. C. Hulbert Jr., G. S. Morgan, &
H. F. Evans. 2008. Terrestrial mammals of
the Palmetto Fauna (early Pliocene, latest
Hemphillian) from the Central Florida
Phosphate District. Pp. 293-312 in X.-M.
Wang & L. G. Barnes, eds. Geology and
Vertebrate Paleontology of Western and
Southern North America, Contributions in
Honor of David P. Whistler. Natural History
Museum of Los Angeles County Science
Series, Number 41.
Webb, S. D., & K. T. Wilkins. 1984. Historical
biogeography of Florida Pleistocene
mammals. Pp. 370-383 in H. H. Genoways
& M. R. Dawson, eds. Contributions in


Quaternary Vertebrate Paleontology: a
Volume in Memorial to John E. Guilday.
Carnegie Museum of Natural History,
Pittsburgh, Special Publication 8.
Witmer, L. M., S. D. Sampson, & N. Solounias.
1999. The proboscis of tapirs (Mammalia:
Perissodactyla): a case study in novel narial
anatomy. Journal of Zoology 249:249-267.
Wu, X. Z., W. Liu, X. Gao, & G. M. Yin. 2006.
Huanglong Cave, a new late Pleistocene
hominid site in Hubei Province, China.
Chinese Science Bulletin 51(20):2493-2499.

APPENDIX 1

Information on Florida Blancan fossil localities that
have produced Tlw ii -,,.. Only approximate location
data provided for sites that are still being collected,
to protect them from vandalism. Most lack formal
stratigraphic names for their sediments, as they are
either isolated sinkhole or alluvial deposits.
SITES WITH TA4PIRUS LUNDELIUSI SP. NOV.
1. Haile 7C (type locality). UF locality AL109, 6
km northeast ofNewberry, Alachua County, Florida
(Morgan & Hulbert 1995:68-69; Emslie 1998).
SE/4 of section 24, T. 9 S., R. 17 E. on USGS 7.5'
Newberry Quadrangle; 29.690 N, 82.56 W. Middle
late Blancan, ca. 1.9-2.2 Ma.
2. Haile 7G. UF locality AL125, 6 km northeast
of Newberry, Alachua County, Florida. -,l of
section 24, T. 9 S., R. 17 E. on USGS 7.5'Newberry
Quadrangle; 29.69 N, 82.560 W. Middle late
Blancan, ca. 1.9-2.2 Ma. Located about 75 m east
of Haile 7G. The lacustrine sediments of the two
localities are possibly continuous, but they are
currently separated by a covered area.
3. Withlacoochee River IA. UF locality MR060,
in channel of Withlacoochee River about 16 km
southeast of Dunnellon, Marion County, Florida.
SE/4 of section 29, T. 17 S., R. 20 E. on USGS 7.5'
Stokes Ferry Quadrangle; 29.00 N, 82.30 W. Middle
late Blancan, ca. 1.9-2.2 Ma. In situ specimens are
collected from sandy gray clay that fills pockets in
the Eocene limestone bedrock.
4. Waccasassa River 9A. UF locality LV040, in






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. f,, )


bank of Waccasassa River about 8 km southeast of
town of Otter Creek, Levy County, Florida. SW'4 of
section 8, T. 14 S., R. 16 E. on USGS 7.5' Bronson
SW Quadrangle; 29.30 N, 82.70 W. Caloosahatchee
Formation (or lateral equivalent), middle late to
latest Blancan, ca. 1.6-2.2 Ma. Vertebrate fossils
collected interbedded with f-p,-e of mollusks
found in the Caloosahatchee Formation in southern
Florida.
5. Inglis lA. UF locality CI001, 2.5 km southwest
of Inglis, Citrus County, Florida. SW/4 of NE /4
of section 9, T. 17 S., R. 16 E. on USGS 7.5'
Yankeetown Quadrangle; 29.0 N, 82.680 W. Latest
Blancan, ca. 1.6-1.9 Ma.
6. Inglis lB. UF locality CI013, 2.5 km southwest
of Inglis, Citrus County, Florida. SW'/4 of NE /4
of section 9, T. 17 S., R. 16 E. on USGS 7.5'
Yankeetown Quadrangle; 29.0 N, 82.68 W. Latest
Blancan, ca. 1.6-1.9 Ma.
7. Inglis ID. UF locality (1 '', 2 km south-
southwest of Inglis, Citrus County, Florida. NW /
of SE/4 of section 10, T. 17 S., R. 16 E. on USGS
7.5' Yankeetown Quadrangle; 29.010 N, 82.670 W.
Latest Blancan, ca. 1.6-1.9 Ma.
8-10. Santa Fe River 8, 8A, and 8C. UF localities
CO006B, CO007B, and C0009B, Columbia
County, Florida. NW/4 of SE14 of section 18, T. 7
S., R. 16 E. on USGS 7.5' Fort White Quadrangle;
29.880 N, 82.670 W. Middle late to latest Blancan,
ca. 1.6-2.2 Ma.
11. Walden Pit 2. UF locality SA050, 8 km east-
northeast of Fruitville, Sarasota County, Florida.
NE/4 of section 23, T. 36 S., R. 19 E. on USGS
7.5' Old Myakka Quadrangle; 27.35 N, 82.370 W.
Pinecrest Formation; early late Blancan, ca. 2.3-
2.6 Ma.
SITES WITH TAPIRUS HAYSII
12-13. Santa Fe River 1 and lB. UF localities
CO003B and C0038B, Columbia County, Florida.
NE'/4 of NW/4 of section 34, T. 7 S., R. 16 E. on
USGS 7.5' High Springs SW Quadrangle; 29.840
N, 82.700 W. Early late Blancan, ca. 2.3-2.6 Ma.
(UF 177841, 177842, 213920, and 224631).
14. Kissimmee River 2. UF locality OB002, 18


km west of Okeechobee, Okeechobee County,
Florida. Section 13, T. 37 S., R. 33 E. on USGS
7.5' Fort Basinger Quadrangle; 27.250 N, 80.97
W. Pinecrest Formation; early late Blancan, ca.
2.3-2.6 Ma. (UF 51250).
15. Kissimmee River 6. UF locality OB006, 18 km
west of Okeechobee, Okeechobee County, Florida.
Section 21, T. 36 S., R. 33 E. on USGS 7.5' Fort
Basinger Quadrangle; 27.33 N, 81.03 W. Pinecrest
Formation; early late Blancan, ca. 2.3-2.6 Ma. (UF
52601). This stretch of the river also produced UF
51249, a p2 of Japirus sp. indeterminate.
16. Inglis IC. UF locality C1019, 2 km south-
southwest of Inglis, Citrus County, Florida. NW'/
of SE/4 of section 10, T. 17 S., R. 16 E. on USGS
7.5' Yankeetown Quadrangle; 29.01 N, 82.670 W.
Latest Blancan, ca. 1.6-1.9 Ma. (UF 177844).
17. De Soto Shell Pit 5. UF locality DEO18, 17 km
south of Arcadia, De Soto County, Florida. NW/4
of SE/4 of section 33, T. 39 S., R. 25 E. on USGS
7.5' Arcadia SE Quadrangle; 27.040 N, 81.820 W.
Caloosahatchee Formation, latest Blancan, ca. 1.6-
1.9 Ma. (UF :"'-', 232075). Neither specimen
was collected in situ, and these two specimens
could be Irvingtonian rather than Blancan, as this
mine also contained sediments from the early
Irvingtonian Bennont Formation.
18. Devil's Elbow 2. UF locality PU005B,
submerged locality on St. Johns River near East
Palatka, Putnam County, Florida. Section 38, T. 10
S., R. 27 E. on USGS 7.5' Hastings Quadrangle;
29.65 N, 81.610 W. Nashua Formation, early late
Blancan, ca. 2.3-2.6 Ma. (UF 177843).
SITES WITH TAPIRUS SPECIES INDETERMINATE
19. Sommers Pit. UF locality SA029, 8 km east-
northeast of Fruitville, Sarasota County, Florida.
NE/4 of section 23, T. 36 S., R. 19 E. on USGS
7.5' Old Myakka Quadrangle; 27.340 N, 82.370 W.
Pinecrest Formation; early late Blancan, ca. 2.3-
2.6 Ma. (UF/TRO 1469).
20. Walden Pit 1. UF locality SA049, 8 km east-
northeast of iw. ,, i il Sarasota County, Florida.
NE/4 of section 23, T. 36 S., R. 19 E. on USGS
7.5' Old Myakka Quadrangle; 27.350 N, 82.37 W.






HULBERT JR.: Blancan Tapirus from Florida


Pinecrest Formation; early late Blancan, ca. 2.3-
2.6 Ma. (UF/TRO 1489).
21. De Soto Shell Pit 2. UF locality DE009, 16 km
south of Arcadia, De Soto County, Florida. NW/
of SW/4of section 27, T. 39 S., R. 25 E. on USGS
7.5' Arcadia SE Quadrangle; 27.050 N, 81.810 W.
Caloosahatchee Formation, latest Blancan, ca. 1.6-
1.9 Ma. (UF 177836).
22. Lehigh Acres Pit. UF locality I I:,., 5.2 km
south of Alva, Lee County, Florida. NE' of section
10, T. 44 S., R. 27E. on USGS 7.5'Alva Quadrangle;
26.670 N, 81.600 W. Caloosahatchee Formation,
late Blancan, ca. 2-2.6 Ma. (UF 162666).
23. Haile 15A. UF locality AL032, 4.7 km northeast
of Newberry, Alachua County, Florida. NW`i of
SW /4of section 24, T. 9 S., R. 17 E. on USGS 7.5'
Newberry Quadrangle; 29.670 N, 82.570 W. Early
late Blancan, ca. 2.3-2.6 Ma. (UF 17468).
24. US 19 Bridge Site. UF locality GI008, north
of bridge over Suwannee River, Gilchrist County,
Florida. SW/4 of SW /4of section 20, T. 10 S., R.
14 E. on USGS 7.5' Fanning Springs Quadrangle;
29.60 N, 82.90 W. Early late Blancan, ca. 2.3-
2.6 Ma. (UF 247180, 248788, 248789, 255357,
255358). This previously unpublished local
fauna also includes ;',, !ia0.. platymarginata,
Pseudemys sp., ,,, n.. sp., Macroclemys sp.,
4:. ,;. ferox, He.sperotestudo sp., -!/ .r sp.,
.D'.'i-[ bellus, the small morph of Holmesina
floridanus, Paramylodon '. h.,is. N1,,in,
peninsulatus, Odocoileus 1;"'. and
Capromeryx arizonensis.
25. Hubbard Pit. UF locality OR003, abandoned
clay and sand quarry, Orange County, Florida.
SW/4 of NW/4 of section 6, T. 22 S., R. 29 E.
on USGS 7.5' Orlando West Quadrangle; 28.60
N, 81.5' W. Late early or early late Blancan, ca.
2.9-2.5 Ma. (UF 2. ,".'). This quarry produced
two fossiliferous horizons with abundant marine
invertebrates and vertebrates. The younger bed,
which also produced a few terrestrial vertebrate
specimens, is of Blancan age based on the presence
of Nannippus peninsulatus and molluscan species
characteristic of bed 11 from Macasphalt Shell Pit
near Sarasota (R. Portell, pers. comm.). In addition


to N. peninsulatus and 7,I' ,; , fossils ofF- '., sp.
and Dasypus bellus were also recovered from this
locality.

APPENDIX 2

List of characters and character states used in the
phylogenetic analyses. They are adapted from
a larger character set originally developed by
Hulbert and Wallace (2005), with phylogenetically
uninformative characters for the group of species
under study here deleted. Original numbering
scheme of characters is retained to maintain
consistency between publications.
3. temporal crests meet to form .. --,'ll. crest: (0) at
or posterior to frontal-parietal suture; (1) anterior
to frontal-parietal suture.
4. dorsal table of frontal. (0) relatively narrow or
small; (1) relatively broad.
6. nasal-frontal lateral profile: (0) nasals and frontal
on same plane; (1) nasals notably stepped down
from frontals.
7. size of interparietal bone: (0) large; (1) small.
8. shape of interparietal bone: (0) typically
polygonal (hexagonal or diamond shaped); (1)
typically triangular.
9. interparietal fusion with occipital: (0) occurs
early in ontogeny (before loss of DP4); (1) occurs
late in ontogeny (after eruption of P4).
10. nasal length: (0) L,,._ longer than width of
combined nasals; (1) short, shorter than width of
combined nasals.
14. fossa for meatal diverticulum on nasal: (0)
shallow and without distinct margins; (1) deep and
with distinct margins.
15. fossa for meatal diverticulum on posterior
dorsal surface of nasal: (0) not extensive, does
not near midline; (1) very extensive, approaches
midline.
16. development of fossa for meatal diverticulum
on dorsal table of frontal: (0) very limited; (1)
broad exposure with distinct posterior margin.
18. posterodorsal process of maxilla widely






BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


exposed dorsal above the orbit, forming base of
trough for meatal diverticulum: (0) yes; (1) no.
20. orientation of lambdoidal crests in adults: (0)
mostly posterior with little or no outward flair; (1)
posterolateral orientation, with notable lateral flair.
22. location of infraorbital foramen relative to the
cheekteeth: (0) dorsal to P4; (1) dorsal to P3 or P2.
25. anteromedial process of maxilla: (0) well
exposed in lateral view dorsal to premaxilla; (1)
covered by premaxilla, not visible in lateral view
or barely so.
26. dorsal maxillary flange: (0) absent; (1) slight;
(2) extensive.
27. length of posterior process of premaxilla: (0)
long, terminates posterior to Pl; (1) terminates
dorsal to or just in front ofP 1; (2) ends about midway
over C-P1 diastema; (3) very short, terminates well
anterior to midway point of diastema.
28. width of maxillary bar between infraorbital
foramen and lacrimal: (0) narrow, usually less than
5 mm; (1) wide, more than 5 mm.
29. shape of lacrimal: (0) narrow, much taller than
long; (1) broad, about as long as it is high.
30. facial surface of lacrimal: (0) flat or convex; (1)
concave.
31. anterior lacrimal processess: (0) absent or very
weak; (1) present, well developed.
33. posterior process of lacrimal: (0) broad and flat;
(1) slender, pointed or knobby.
34. typical number of large lacrimal foramen: (0)
two; (1) one.
37. relative diastema length: (0) short (ldl/p2m3L
< 0.40); (1) medium (0.40 < ldl/p2m3L < 0.50); (2)
long (ldl/p2m3L > 0.50).
38. location of mental foramen: (0) anterior to the
p2; (1) directly ventral to the p2.
39. orientation of anterior margin of ascending
ramus of mandible in lateral view: (0) projects
vertically and posteriorly, not anteriorly (does not
overlie the m3); (1) projects anteriorly as ,-.11 as
vertically below the coronoid process, often lies
dorsal to m3.


40. depth of horizontal ramus below m3: (0)
shallow (m3depth < 18% mandL); (1) moderate
(m3depth between 18 and 19% of mandL); (2)
deep (m3depth between 19 and -'..,, of mandL);
(3), very deep (m3depth > 20% of mandL).
41. relative crown height of cheek teeth: (0) short,
relatively brachydont; (1) taller.
43. PI with: (0) single, small posterolingual cusp
(=hypocone of some) and lingual, in -iiiiir but no
cross lophs or other cusps; (1) large posterolingual
cusp, sometimes with accessory cusps and often
with some development of a transverse loph; (2)
large posterolingual cusp with strong, complete
transverse loph; (3) no distinct posteriorlingual
cusp, only a cingulum.
45. P2 AW/PW ratio: (0) on average, less than or
equal to 0.85; (1) on average, greater than 0.85.
47. transverse lophs on P2-P4 and p2-p4: (0) poorly
separated; 1; well separated.
48. P2 protoloph: (0) does not reach ectoloph; (1)
just reaches to base of ectoloph; (2) merges midway
or higher onto ectoloph.
49. P3 protoloph: (0) does not reach ectoloph; (1)
just reaches to base of ectoloph; (2) merges midway
or higher onto ectoloph.
51. P2 metaloph: (0) does not reach ectoloph; (1)
just reaches to base of ectoloph; (2) merges midway
or higher onto ectoloph.
55. parastyle development on P3-M3: (0) moderate;
(1) strong; (2) very strong.
56. labial cingulum on posterior half of upper
cheek teeth: (0) present on half or more of P3-M3;
(1) absent or very rare on P3-M3.
67. il morphology and size: (0) il slightly larger
than i2, not procumbent; (1) il larger than i2,
slightly procumbent; (2) il much larger than i2,
very procumbent and spatulate.
69. length of p2 relative to p3: (0) short, (p2 L)/(p3
L) less than 1.1; (1) long, this ratio greater than 1.1.


72. relative
hypolophid
hypolophid)


height of unworn protolophid and
(or hypoconid/entoconid if no
on premolars: (0) protolophid






HULBERT JR.: Blancan Thpirus from Florida 125


distinctly taller than hypolophid; (1) the two are
approximately equal in height.
73. cristid obliqua on p3: (0) strong, blocks
interlophid valley labially; (1) very weak or absent.







BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. ',, )


APPENDIX 3.
Character state matrix used in phylogenetic analyses. Character numbers in top row correspond to those
in Appendix 2. A"?" indicates a unknown or non applicable character state.


Outgroup
T: hezhengensis
T johnsoni
7. webbi
T. polkensis
T lundeliusi
7. haysii
T. veroensis
T bairdii
T terrestris
T. pinchaque





Outgroup
T hezhengensis
T. johnsoni
7: webbi
. j ...., .. .
T lundeliusi
7: havsii
T veroensis
7. bairdii
T. lerrestris
T i '- ..,


4 6 7 8 9 10 14 15 16 18 20 22 25 26 27 28 29 30 31
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
? ? ? ? ? ? ? ? ? ? ? 0 0 1 ? ? ? ? ?


0 0 0
0 0 0
0 0 1
0 0 1
0 0 1
0 0 1
1 0 1
1 1 ?
0 1 0



37 38
0 0


1 0
1 1
0 1
0 1
0 1
0 1
2 1
1 1
1 1


? 0
? 0
1 0
0 0
1 1
1 1
0 0
? 0
0 0



39 40
0 0

? 9
0 2
0 0
0 1
1 3
1 3
1 2
0 1
0 1
0 0


0 0 0
0 0 ?
0 0 0
0 1 0
1 1 0
1 1 0
0 0 0
0 0 1
0 0 1


43 45
0 0

0 0
1 0
1 1
0 1
1 1
2 1
1 1
0 1
1 1
0 1


55 56 67 69 72 73
0 0 0 0 0 0

0 0 ? ? 9 ?
0 0 1 0 0 0
1 0 2 0 1 0
1 0 2 0 1 0
2 1 2 1 1 0
2 1 2 1 1 1
2 1 2 1 1 1
1 0 2 0 1 ?
1 1 2 0 1 1
1 0 2 0 1 0






The BULLETIN OF THE FLORIDA MUSEUM OF NATURAL HISTORY publishes original research
conducted by our faculty, staff, students, and research associates. We also accept submissions of appropriate,
fully funded m,!,n,-..-.il., from external researchers. Priority is given to monograph-length -'.'-"-*-. describing
specimen-based research. Forty-five double-spaced pages (including references and appendices but excluding
figures, tables, and supplemental materials) is the minimum length for submitted manuscripts, although there can
be exceptions as determined by the Managing Editor. Starting in 2010, the Bulletin is published simultaneously
in two formats. Approximately 400 printed copies are distributed to libraries and museums world-wide by the
University of Florida Library system. Authors have the option of purchasing additional printed copies at cost for
distribution to colleagues and associates. An identical, electronic version is posted in PDF format on the Florida
Museum of Natural History web site simultaneously with the availability of the printed version and is available
free of charge for reading or downloading. Supplemental materials are available only through the web site.

INSTRUCTIONS FOR AUTHORS

Detailed instructions are posted on the web site http://www.flmnh.ufl.edu/bulletin/bulletin_inst.htm. All taxonomic
papers must adhere to the rules published in the appropriate international code of systematic nomenclature.

RECENT PUBLICATIONS OF THE FLMNH BULLETIN

Thomson, A. W., and L. M. Page. 2010. Taxonomic revision of the Amphilius uranoscopus group (Teleostei:
Siluriformes) in Kenya, with the description of a new species from the Athi River. 49(2):45-66. $7.00

Franz, R., & S. E. Franz. 2009. A new fossil land tortoise in the genus ( ;'', ,. i. '... (Testudines: Testudinidae)
from the northern Bahamas, with an osteological assessment of other Neotropical tortoises. 49(1):1-44. $8.00

Snelson, F. F., Jr., T. J. Krabbenhoft, & J. M. Quattro. 2009. Elassoma ;il.' n,, a new _.,"-p.-- -. of pygmy sunfish
(Elassomatidae) from Florida and Georgia. 48(4):119-144. $7.00

Zaspel, J. M., S. J. Weller, & R. T. Carde. 2008. A review of Virbia (formerly Holomelina) of North America.
48(3):59-118. $10.00

Webb, S. D. 2008. Revision of the extinct Pseudoceratinae (Artiodactyla: Ruminantia: Gelocidae). 48(2):17-58.
$7.00

Steadman, D. W. 2008. Doves (Columbidae) and cuckoos (Cuculidae) from the early Miocene of Florida. 48(1): 1-
16. $7.00

Borden, W. C., & M. M. CG-ill_.u 2008. Striated muscles of the black basses (I f,-l ?i,-", Centrachidae):
myological stasis in a generalized group of Percomorph fishes. 47(4): 109-136. $7.00

Wilkins, L., J. M. Allen, & D. Reed. 2007. Methods of assessing health and diet of Florida panthers (Puma
concolor) using museum specimens, parts 1 & 2. 47(3):73-108. $7.00

Ehret, D. J. 2007. Skeletochronology: A method for determining the individual age and growth of modern and
fossil tortoises (Reptilia: Testudines). 47(2):49-72. $7.00

A complete list of available issues and current prices of the Bulletin of the Florida Museum of Natural History
and instructions for purchasing them can be found at http://www.flmnh.ufl.edu/bulletin/bulletin_vols.htm. Digital
versions of out-of-print issues can be found at http://ufdcwebl.uflib.ufl.edu/ufdc/?a=ufirg&m=hitflmnh.







Full Text

PAGE 1

BULLETIN A NEW EARLY PLEISTOCENE TAPIR (MAMMALIA: PERISSODACTYLA) FROM FLORIDA, WITH A REVIEW OF BLANCAN TAPIRS FROM THE STATE Richard C. Hulbert Jr. Vol. 49, No. 3, pp. 67 2010 UNIVERSITY OF FLORIDA GAINESVILLE

PAGE 2

The FLORIDA MUSEUM OF NATURAL HISTORY is Floridas state museum of natural history, dedicated to understanding, preserving, and interpreting biological diversity and cultural heritage. The BULLETIN OF THE FLORIDA MUSEUM OF NATURAL HISTORY is a peer-reviewed journal that publishes results of original research in zoology, botany, paleontology, archaeology, and museum science. The Bulletin is published at irregular intervals, and volumes are not necessarily completed in any one year. Volumes contain between 150 and 300 pages, sometimes more. The number of papers contained in each volume varies, depending upon the number of pages in each paper, but four numbers is the current standard. Multi-author issues of related papers have been published together, and inquiries about putting together such issues are welcomed. Address all inquiries to the Managing Editor of the Bulletin. Jacqueline Miller, Guest Editor for this number Richard C. Hulbert, Jr., Production Editor for this number Bulletin Committee Ann S. Cordell Richard C. Hulbert, Jr. Jacqueline Miller Larry M. Page Roger W. Portell, Treasurer Irvy R. Quitmyer David W. Steadman, ISSN: 0071-6154 Copyright 2010 by the Florida Museum of Natural History, University of Florida. All rights reserved. Text, commercial use or republication by printed or electronic media is strictly prohibited without written permission of the museum. Publication Date: September 30, 2010 Send communications concerning puchase or exchange of this publication and manuscript queries to: Managing Editor of the Bulletin Florida Museum of Natural History University of Florida P.O. Box 117800 Gainesville, FL 32611-7800 USA FAX: 352-846-0287

PAGE 3

ABSTRACT Florida has the best late Blancan (early Pleistocene, 2.6.6 Ma) record of Tapirus in North America. The genus is currently known from 25 fossil localities of Blancan age, ranging through most of the peninsular region of the state. Three species are recognized from this time interval in Florida: the relatively large-sized Tapirus haysii Leidy; the moderate-sized Tapirus lundeliusi sp. nov. ; and a much smaller, as yet unnamed species. Tapirus haysii was previously known from the early Irvingtonian (ca. 1.6 to 1.0 Ma) in Florida; the specimens reported here from seven localities extend its range into the Blancan, which is in accord of T. haysii in Florida are all relatively sparse, mostly consisting of one or two isolated teeth per locality. The unnamed small Blancan species of Tapirus is slightly better known: three jaw fragments each with Tapirus and all of those previously described from the Pleistocene of North America. Dwarf tapirs are known from the Hemphillian of the southeastern United States (e.g., Tapirus polkensis ) and the late Miocene of Europe (e.g., Tapirus pannonicus ), but this is the Tapirus polkensis and is the same size as an undescribed species of Tapirus present in the late Hemphillian Palmetto Fauna of central Florida. Tapirus lundeliusi sp. nov. is described on the basis of multiple skeletons from the Haile 7C and 7G localities of north-central Florida, and more limited samples from nine other Blancan sites. It apparently 1A locality. Tapirus lundeliusi is of moderate size for the genus, similar to extant Tapirus terrestris and Tapirus pinchaque and the middle to late Pleistocene species Tapirus veroensis than Tapirus haysii The skull of T. lundeliusi differs from those of T. veroensis and T. haysii in retaining a bone, and ontogenetically rapid fusion of interparietal to occipital. Derived features of T. lundeliusi include a reduced frontal shield and increased relative mandibular condyle height. Phylogenetic analysis places T. lundeliusi as the sister taxon of T. veroensis + T. haysii and the clade formed by these three North American fossil species is formally designated as Tapirus ( Helicotapirus ) subgen. nov. Among other species of Tapirus the extant T. bairdii and the late Neogene T. polkensis are most closely related to T ( Helicotapirus extensive meatal fossa on the dorsal surface of the frontal and nasal bones. Key Words: Tapiridae; Tapirus ; new species; Blancan; Florida; phylogenetic analysis; biochronology A NEW EARLY PLEISTOCENE TAPIR (MAMMALIA: PERISSODACTYLA) FROM FLORIDA, WITH A REVIEW OF BLANCAN TAPIRS FROM THE STATE Richard C. Hulbert Jr. 1 Hulbert Jr., R. C. 2010. A new early Pleistocene tapir (Mammalia: Perissodactyla) from Florida, with a review of Blancan tapirs from the state. Bulletin of the Florida Museum of Natural History 49(3):67. 1

PAGE 4

68 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) INTRODUCTION F ossils referred to the extant genus of tapir, Tapirus Brisson 1762, are relatively abundant in Irvingtonian and Rancholabrean faunas in Florida, and have received considerable study (e.g., Sellards 1918; Simpson 1929 1945; Bader 1957; Ray 1964; Lundelius & Slaughter 1976; Ray & Sanders 1984; Hulbert 1995; Graham 2003). Two species are currently recognized from Florida during this interval, Tapirus haysii Leidy in the early Irvingtonian and Tapirus veroensis Sellards in the late Irvingtonian through Rancholabrean (Hulbert 1995). The situation is quite different for the preceding Blancan land mammal age (4.4.6 Ma) in Florida, for which there are no formally described species of Tapirus presence of Tapirus in the Blancan of Florida, listing records in a table from three localities, Santa Fe River 1B, Haile 15A, and Inglis 1A (Fig. 1; Appendix 1), but no descriptions were Ray et al. (1981) and MacFadden and Hulbert (2009) noted that Webbs (1974) list of Santa Fe River 1B mammals is actually a composite listing of Blancan mammals (or those thought at the time to be Blancan) from all Santa Fe River fossil sites, not just the 1B locality. Furthermore, as detailed by MacFadden and Hulbert (2009), the Santa Fe River 1B locality actually produced both Blancan and Rancholabrean vertebrates, contrary to the assertions of Webb (1974), who was followed by Kurtn and Anderson (1980), that site contained only Blancan fossils. Two size classes of Tapirus have been found in the Santa Fe River 1 region (Fig. 1): a rarer, larger form that is within the size range of Tapirus haysii and a much more numerous group of smaller teeth and bones that is within the size ranges of both the Rancholabrean Tapirus veroensis and the new Blancan species described below. The Santa Fe River 1 and 1B specimens of T haysii are all isolated TABLE OF CONTENTS Introduction ........................................................................ 68 Materials and Methods ....................................................... 72 Fossil localities and Florida Blancan Biochronology ........ 76 Systematic Paleontology .................................................... 78 Helicotapirus subgen. nov. ......................................... 78 Tapirus lundeliusi sp. nov. .......................................... 80 Tapirus haysii Leidy 1859 .......................................... 81 Tapirus subgenus and species indeterminate .............. 86 Description of Tapirus lundeliusi ....................................... 87 Description and Comparisons of Tapirus sp. from Sommers Pit, Walden Pit 1, Kissimmee River 6, Hubbard Pit, and US 19 Bridge Site ......................... 109 Phylogenetic Relationships .............................................. 111 Discussion ........................................................................ 113 Conclusions ...................................................................... 116 Acknowledgements .......................................................... 116 Literature Cited ................................................................ 117 Appendix 1 ....................................................................... 121 Appendix 2 ....................................................................... 123 Appendix 3 ....................................................................... 126

PAGE 5

HULBERT JR.: Blancan Tapirus from Florida 69 show that T. haysii was present in the late Blancan of Florida (contrary to Hulbert 1995). The smaller teeth of Tapirus from the Santa Fe River 1 region probably represent a mixture of Rancholabrean T. veroensis and the new Blancan species, but most isolated teeth of the two species can not be standard linear measurements. Linear discriminant analyses between the two taxa for each cheektooth 13 to over 45 percent, with most falling in the range of 25 to 30 percent. Discriminant scores for most Santa Fe River 1 teeth grouped them with T veroensis as the Blancan species. However, a hypothesis that they are all T veroensis reason they are not included in this study. Figure 1. Map of Florida showing location of fossil localities of Blancan age that have produced fossils of Tapirus See Appendix 1 for detailed geographic information. Tapirus lundeliusi sp. nov. was found at localities numbered 1 through 11; Tapirus haysii at localities 12 through 18; and Tapirus sp. indeterminate at localities 15 and 19 through 25.

PAGE 6

70 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) There are no problems with mixing or reworking of specimens from the two other Blancan localities with Tapirus listed by Webb (1974), Haile 15A and Inglis 1A. However, the single Haile 15A specimen, UF 17468, is a fragmentary lower cheektooth. As noted by Robertson (1978), it is adequate to document the presence of the genus in 1A sample of Tapirus consists primarily of isolated postcranial elements, with a few isolated teeth, two dorsal skull elements, and three partial maxillae. This sample was also listed by Webb and Wilkins (1984) and Graham (2003), but has never been described or Two skeletons of Tapirus a subadult, UF 121736, and a juvenile, UF 162351, were found the late Blancan Haile 7C locality in 1989 tapirs older than late Pleistocene ever found in of Blancan tapirs from the state. The skull of UF from Tapirus veroensis Following its preparation in the early 1990s, it was evident that it represented a new species. However, formal description was delayed for several reasons including, the lack of a complete nasal bone in UF 121736 (which provide Figure 2. Teeth of Tapirus haysii from the late Blancan of Florida, in occlusal view. A UF 213920, left M1 or M2; B UF 224631, right M1 or M2; C UF 177842, left p4; D UF 51250, left P4; E UF 177843, right M3; F UF 177844, left P1. Localities listed in text and Appendix 1.

PAGE 7

HULBERT JR.: Blancan Tapirus from Florida 71 Figure 3. Portions of an articulated juvenile skeleton of Tapirus lundeliusi sp. nov., UF 247102, from Haile 7G. Other portions of this skeleton, including remainder of forelimbs and both hind limbs removed prior to photography. Degree of association of the tapir skeletons from Haile 7C and 7G ranged from nearly fully articulated, such as this individual, to those scattered over an area of several square meters with little or no direct articulation. Abbreviations: C, cervical vertebra; hum, humerus; L, left (unless followed by numeral); L (followed by numeral), lumbar vertebra; mand, mandible; max, maxilla; R, right; scap, scapula; T, thoracic vertebra.

PAGE 8

72 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) key characters in this genus), the relative young ontogenetic ages of the two specimens (which therefore might not display fully adult character states for some features), and the low sample size which greatly limited quantitative comparisons with other species. S. D. Emslie led a second excavation at Haile 7C in December 1994January 1995, with the of fossil birds (see Emslie 1998). The third Haile was found by volunteer Reed Toomey early in this dig. Although missing many of the bones of the manus and pes, UF 160715 is otherwise extremely complete and well preserved. Its skull preserved the nasal bones, unlike that of UF 121736. Several partial skeletons of tapirs consisting of only postcranial elements were also found during the 1994 excavation. Morgan and Hulbert (1995) noted the presence of a new species of tapir from Haile 7C in a brief description of the site. Three additional tapir skeletons with skulls, two adults and a juvenile, were collected during the last major The quantity of known Blancan tapir specimens in Florida changed dramatically with the discovery of the Haile 7G locality in the Spring of 2005. Located less than 100 m east of Haile 7C (Fig. 1), Haile 7G had very similar sediments (laminated blue-gray clays mixed with limestone boulders and interspersed with lenses and thin layers of sand-size fragments of limestone), also frequently produced associated to fully articulated skeletons (Fig. 3), and shared most of the fauna of its neighbor (Table 1). While proximity of vertebrate fossil localities in the Haile limestone mining region of Florida does not necessarily mean that the sites will be the same age, it is the case for Haile 7C and 7G. But the Haile 7G locality was of greater size and produced vastly more vertebrate skeletons. From May 2005 to April 2008 we collected over 500 tetrapod skeletons at Haile 7G, and Tapirus was the most common mammal at the site, represented by over 70 partial to nearly complete skeletons, all representing the same species found at Haile 7C. As of 2010, about half of the Haile 7G skulls and mandibles had been prepared and are available for study, and many of these are juveniles. But the combined sample from Haile 7C and 7G, augmented by those from Inglis 1A, Santa Fe 8, and elsewhere, is exceeded among fossil tapirs in terms of overall numbers of skeletons and large samples of all ontogenetic stages of skeletal and dental development only by that of Tapirus polkensis (Hulbert et al. 2009). The purposes of this study are to formally name and describe this new species of Tapirus concentrating on characters of the skull and dentition, and to place it in a phylogenetic context relative to Tapirus veroensis Tapirus haysii and other species in the genus. A comprehensive quantitative study of the Haile 7G sample (including postcranial skeletal elements) will be done upon its complete preparation and curation. Ecological analyses using stable oxygen and carbon isotopes of the Haile 7G sample of Tapirus are on-going (e.g., De Santis & MacFadden 2008). Other late Blancan species of Tapirus in Florida are also MATERIALS AND METHODS All measurements are reported in millimeters. Measurements and general methods are the same as described in Hulbert (2005) and Hulbert et al. (2009). To the comparative samples of Tapirus veroensis used in Hulbert (2005), I have added specimens accessioned into the UF collection since 2005, most notably a partial mandible (UF 225215) and isolated teeth (UF 225221) from Steinhatchee River 2, and partial mandibles from Jacksonville Beach (UF 223228), Bedman Creek (UF 223709), Peace River (UF 227141), and Dickerson Coquina Pit (UF 244022). The Hulbert (2005) sample of Tapirus haysii has been augmented by recently donated specimens from Leisey Shell Pit 3 (UF 241860, 241870, 241874, 241948, 242001, 243500). The opinion of Ray and Sanders (1984) that T. haysii is a valid name and the senior synonym of Tapirus copei Simpson is followed here (see also Hulbert 1995, 2005; Sanders 2002; Graham 2003). Dalquest and Schultz (1992) and Smith and Cifelli (2000) promoted the continued

PAGE 9

HULBERT JR.: Blancan Tapirus from Florida 73 use of T. copei instead of T. haysii but without citing any evidence, morphologic or otherwise, to disprove the convincing argument made by Ray and Sanders (1984). The recent decision by the International Commission on Stratigraphy to move the Gelasian Age to the Pleistocene Epoch (Gibbard et al. 2010), and thus changing the beginning of the Quaternary (and Pleistocene) from about 1.81 to 2.58 Ma, is followed here. Boundaries and subdivisions of North American land mammal ages generally follow Bell et al. (2004), with the Blancan/Irvingtonian boundary at about 1.6 Ma and the Irvingtonian/ Rancholabrean boundary at about 0.3 Ma. Thus the later portion of the Blancan land mammal age now falls within the early Pleistocene, not the late Pliocene. All of the Florida fossil localities under study here fall in the later portion of the Blancan, within the interval of about 2.6 to 1.6 Ma. As in Hulbert (2005) and Hulbert et al. (2009), the genus Tapirus a closer common ancestry with the type species of the genus, Tapirus terrestris (Linnaeus), than with the type species of all other valid genera in the Phylogenetic analyses were done using a set of 22 cranial, four mandibular, and 13 dental characters (Appendix 2) for nine species of New World Tapirus (Appendix 3). The late Miocene Chinese species Tapirus hezhengensis was included in one analysis to determine its effect on the relationships between New World Tapirus Character states for T. hezhengensis were taken from Deng et al. (2008). To overcome problems with incompleteness, a composite outgroup was made using states from the four best known early Miocene tapirids, Miotapirus harrisonensis Schlaikjer, Nexuotapirus marslandensis (Schoch & Prins), Plesiotapirus yagii (Matsumoto), and Paratapirus helveticus (von Meyer). The character state matrix is shown in Appendix 3. Most parsimonious trees were computed by PAST, version 1.91, using the branch-and-bound algorithm and separate analyses with ordered and unordered multi-state characters. I NSTITUTIONAL A BBREVIATIONS UF Florida Museum of Natural History, University of Florida, Gainesville, Florida, USA; UF/TRO Timberlane Research Organization collection, now housed at the Florida Museum of Natural History; USNM U.S. National Museum, Smithsonian Institution, Washington, D.C., USA. M ORPHOLOGICAL A BBREVIATIONS L greatest length; DL postcanine diastema length; HT height; AW greatest anterior width measured across the protoloph or protolophid near the base of the crown; PW greatest posterior width measured across the metaloph or hypolophid; W width; i lower incisor; I upper incisor; c lower canine; C upper canine; m lower molar; M upper molar; p lower premolar; P upper premolar (a numeral following a tooth abbreviation indicates molar). A d or D in front of a tooth abbreviation indicates a deciduous tooth (e.g., DP1). S TATISTICAL A BBREVIATIONS x sample mean; s sample standard deviation; MIN minimum value observed in a sample; MAX maximum value observed in a sample; OR observed range of a sample; N sample size; CV p probability. O NTOGENETIC D ESCRIPTORS Hulbert et al. (2009) designated seven general ontogenetic life history stages in Tapirus based on tooth eruption and wear. Their temporal durations are not equal, but their use allows comparisons between ontogenetically equivalent individuals. Very young juvenile : DP1DP3 and dp2dp3 with little or no wear are the only fully erupted cheekteeth; DP4 and dp4 may be erupting. Young juvenile : DP1DP4 and dp2dp4 are all fully erupted; M1 and m1 may be erupting; teeth slightly worn. Juvenile : DP1M1 and dp2m1 are all fully erupted and in wear. Crowns of adult premolars and second molars fully formed in crypts. Subadult : P1P3, DP4, M1 and p2p3, dp4, m1 are fully erupted and in wear; M2 and m2 may be erupting; little or no wear on P1P3 and p2p3, heavy wear on DP4 and dp4. Young Adult : P4, M2, p4 and m2 have erupted and are in wear; M3 and m3 either erupting or

PAGE 10

74 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Table 1. Mammals present at Florida late Blancan (early Pleistocene) vertebrate fossil localities with Tapirus lundeliusi sp. nov. (excluding incompleteness of available specimens or absence of needed comparisons with related species. After Morgan and Hulbert (1995), Morgan and White (1995), Ruez (2001), Hulbert et al. (2006), Hulbert and Morgan (2009), and Tedford et al. (2009). Walden Pit 2 Haile 7C Haile 7G With 1A Wacc 9A SFR 8 SFR 8A SFR 8C Inglis 1A Inglis 1B Inglis 1D Dasypus bellus X X X X X X Pachyarmatherium leiseyi X X X X X X X Glyptotherium arizonae X X X X X X X Megalonyx leptostomus X X X X X Paramylodon harlani X X X X X X X ? Eremotherium eomigrans X X X Canis edwardii X ? X Canis lepophagus ? Urocyon citrinus X Procyon rexroadensis X X ? Procyon lotor X X ? ? Arctodus pristinus ? X X Trigonictis macrodon X X X X X Satherium piscinarium ? X Lontra canadensis X Mustela frenata ? Spilogale putorius X X Lynx rufus X X Lynx rexroadensis X ? X X Miracinonyx inexpectata X X X Smilodon gracilis X X X Xenosmilus hodsonae X X ? X Chasmaporthetes ossifragus X Scalopus aquaticus ? X X X X Blarina carolinensis X Cryptotis parva ? ? X Sylvilagus webbi ? X X X X ? X ? X Lepus sp. X

PAGE 11

HULBERT JR.: Blancan Tapirus from Florida 75 Table 1. Continued. Walden Pit 2 Haile 7C Haile 7G With 1A Wacc 9A SFR 8 SFR 8A SFR 8C Inglis 1A Inglis 1B Inglis 1D Aztlanolagus n. sp. X X Sciurus carolinensis ? X ? Glaucomys n. sp. X Castor californicus X X Orthogeomys propinetis X X X X X X X Baiomys sp. X Peromyscus hagermanensis ? Peromyscus sarmocophinus X X Reithrodontomys wetmorei ? X X Neotoma sp. X X Sigmodon medius X X Sigmodon curtisi X X Ondatra idahoensis X X X Erethizon poyeri X X X Erethizon kleini X X Neochoerus dichroplax X X X X X X Tapirus lundeliusi X X X X X X X X X X X Nannippus peninsulatus X X Cormohipparion emsliei X Equus sp. X X X X X X X X Platygonus bicalcaratus X X X ? Hemiauchenia macrocephala ? ? X X X X X X X X Hemiauchenia gracilis X X X X X X Odocoileus virginianus X X X X X Capromeryx arizonensis X X X X X X X Mammut americanum X Cuvieronius tropicus X ?

PAGE 12

76 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) fully erupted, but with little or no wear. Full Adult : Lophs of M3 and m3 moderately worn, but no exposed dentine. Old Adult : Lophs of M3 and m3 have exposed dentine because of heavy wear. FOSSIL LOCALITIES AND FLORIDA BLANCAN BIOCHRONOLOGY The specimens of the new tapir species were collected from ten different localities in central Florida and one in southwestern Florida (Fig. 1; Table 1; Appendix 1). Haile 7C and Haile 7G produced the majority of specimens. Both carbonate sand, and boulders that are interpreted to be remnants of a once much larger and deeper sinkhole lake (Morgan & Hulbert 1995; Emslie 1998; Hulbert et al. 2006; Morgan & Emslie 2010). In addition to sedimentology (Shunk et al. 2006), this hypothesized depositional environment is supported by the frequent preservation of complete or nearly complete vertebrate skeletons (a very rare occurrence for Florida karst fossil sites; Fig. 3) frogs, turtles, alligator, and waterfowl. Tapirs were the most common large mammal at both Haile 7C and 7G. Even though Pleistocene sites in Florida are renown for their relative abundance of tapirs (e.g., Ray & Sanders 1984), they are typically far outnumbered by other medium-sized ungulates such as Equus Palaeolama Hemiauchenia Odocoileus and/or Platygonus The extraordinary abundance of tapirs at Haile 7C and 7G is likely a combination of a vast area of their preferred forested habitat (DeSantis & MacFadden 2007) to generate a large standing population plus an as yet undetermined taphonomic condition to produce the concentrations of skeletons (Hulbert et al. 2006). The fossils of the new species of Tapirus from the remaining nine localities (Appendix 1) are less complete and numerous than at Haile 7C and 7G, and thus serve primarily to supplement the Haile samples and provide information on the chronologic range of the new species. Withlacoochee River 1A is a previously unreported submerged locality (Fig. 1). Fossils were recovered over a 30 m stretch of the river channel, both lying loose on the surface and excavated directly from shallow pockets of gray sandy clay in the Eocene limestone bedrock. The mammalian fauna has a high diversity of xenarthrans, carnivorans, and ungulates (Table 1). The single tapir tooth recovered from this locality, a partial M3, is slightly larger than any found at Haile 7C or 7G. But it is much smaller than M3s of T haysii and within the size range of T veroensis Because the latter contains a much larger sample size than currently known for the new Blancan species, this suggests that the potential size range of the new species should include specimens as large as the Withlacoochee River 1A tooth. Waccasassa River 9A (Fig. 1; Table 1) is another previously unpublished Blancan local fauna. Vertebrate fossils were collected from a river bank in association with a molluscan fauna equivalent to that of the Caloosahatchee Formation in southwest Florida. Fossils of tapirs were collected both from the riverbed in the general Santa Fe River 8 area and area, the ones designated Santa Fe River 8A and 8C (Fig. 1; Table 1). In contrast with the fossiliferous deposits from the much more widely studied Santa Fe River 1 region (Morgan & Hulbert 1995; MacFadden & Hulbert 2009), those from Santa Fe River 8 are apparently in situ early Pleistocene sites, and not late Pleistocene or Holocene sites containing reworked Blancan fossils. The only prior published description of a Santa Fe 8 specimen was in Ray et al. (1981). Fifteen separate fossiliferous karst deposits about 2 km southwest of the town of Inglis (Fig. 1; Appendix 1) were exposed during excavations for the Cross Florida Barge Canal. Of these, only four produced fossils of Tapirus T. haysii at Inglis 1C (Ruez 2001), and the new species at Inglis 1A, 1B, and 1D. Inglis 1A has the largest and most diverse fauna of the Inglis sites (Table 1), and has the best sample of the new tapir excluding the two Haile sites. Members of its fauna have been the subjects of numerous publications (e.g., Meylan 1982; Morgan & Hulbert 1995; Emslie 1998; Tedford et al. 2009; Morgan & Emslie 2010). Inglis 1B and 1D

PAGE 13

HULBERT JR.: Blancan Tapirus from Florida 77 are two of the lesser know Inglis sites. Their faunas are generally a subset to that of Inglis 1A, with a few exceptions, such as the presence of Mylohyus at Inglis 1B and Palaeolama at Inglis 1D. Macasphalt Shell Pit was an extensive commercial sand and shell quarry in Sarasota County that produced well studied invertebrate and vertebrate fossil faunas (e.g., Petuch 1982; Hulbert 1988; Jones et al. 1991; Emslie 1992). The late Blancan land vertebrate assemblage associated with the Pinecrest mollusks from this mine is most similar to the Haile 15A and Santa Fe River 1 local faunas from northern Florida (Morgan & Hulbert 1995). No fossils of tapirs were recovered at Macasphalt Shell Pit. However, Pinecrest mollusks were also mined during the 1970s from two nearby, much smaller quarries, Sommers Pit and Walden Pit (Fig. 1; Appendix 1). Vertebrate taxa from Sommers Pit diagnostic of the late Blancan include Trachemys platymarginata (Weaver & Robertson), the very small morph of (Robertson), Neochoerus dichroplax Ahearn and Lance, Nannippus peninsulatus (Cope), and Cormohipparion emsliei Hulbert. Two separate quarries were mined at Walden Pit, here designated as the more western Pit 1 and more eastern Pit 2, separated by less than 100 m. They are about 1 km north of Sommers Pit. Fossils of terrestrial mammals were uncommon at both quarries, but Walden Pit 1 produced specimens of N. dichroplax and N. peninsulatus while the very small morph of C. emsliei and N. peninsulatus were recovered from Pit 2. Specimens of T. platymarginata were common at both pits. Walden Pits 1 and 2 and Sommers Pit each produced only a single specimen of Tapirus comprising an isolated lower premolar and two mandibular fragments with one tooth each. Despite the very limited nature of these samples, they are important for two reasons. One of the jaw fragments is the only referred specimen of the new species from the southern half smaller than most other known Blancan Tapirus most likely a holdover of a Hemphillian species, and thus bring the total number of late Blancan species of Tapirus in Florida to three. This small taxon is also present at Kissimmee River 6, where it co-occurs with Tapirus haysii Hubbard Pit, and the US 19 Bridge Site (Appendix 1). The latter two are also previously unpublished late Blancan localities. Morgan and Hulbert (1995) proposed that late Blancan, Irvingtonian, and Rancholabrean vertebrate faunas in Florida could be grouped into eight biochronologic assemblages. These are still valid, although new discoveries have changed Irvingtonian by Bell et al. (2004), the earliest Irvingtonian assemblage of Morgan and Hulbert (1995:85) is now the latest interval of the Blancan. Morgan and Hulberts (1995) second assemblage was based solely on the Haile 7C local fauna and subsequent recovery of the richer Haile 7G and Withlacoochee River 1A local faunas (Table 1) allows for a much better biochronologic separation of Florida late Blancan vertebrate faunas into three distinct assemblages, with approximate numeric ages of 2.2.6 Ma for the early late Blancan, 1.9 2.2 Ma for the middle late Blancan, and 1.6.9 Ma for the latest Blancan (Hulbert & Morgan 2009). The early late Blancan interval is characterized by the presence of Paramylodon garbanii (Montellano & Carranza-Castaeda) (proper name for Florida specimens formerly referred to the South American species Glossotherium chapadmalense Kraglievich; see Morgan 2008 for taxonomy), frequent occurrence of the hipparionine equids Nannippus and Cormohipparion and the very small form of Holmesina (as discussed by Hulbert & Morgan 1993). Tapirus is typically rare or absent at early late Blancan sites in Florida. Records of Tapirus haysii are limited to small samples of isolated teeth at Santa Fe River 1 sites, Kissimmee River 2, and Devils Elbow, and a metacarpal 5 from Kissimmee River 6 (Fig. 2; Appendix 1). Fossils from Sommers Pit, Walden Pits 1 and 2, and the US 19 Bridge Site also fall within this time interval. All records of the small, unnamed species of Tapirus belong to this subinterval of the late Blancan; no younger records

PAGE 14

78 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) are currently known. Haile 15A and Lehigh Acres Pit each produced a single partial tooth diagnostic only to the family level, but they presumably represent Tapirus as it is the only known genus in North America from the late Miocene to the Pleistocene. The middle late Blancan interval in Florida Paramylodon harlani (Owen), which replaced the much smaller P garbanii of the preceding interval. Other species the middle late Blancan are Canis edwardii Gazin, Sciurus carolinensis Gmelin, Reithrodontomys wetmorei Hibbard, Ondatra idahoensis Wilson Erethizon kleini Frasier, and Erethizon poyeri Hulbert The last occurrence of Sigmodon medius Gidley occurs during or at the end of this interval. Specimens of are assemblage (Hulbert & Morgan 1993). Haile 7C, Haile 7G, and Withlacoochee River 1A are the primary examples of middle late Blancan faunas. The Waccasassa River 9A and Santa Fe River 8 localities also likely belong in this interval, but they lack the necessary taxa to differentiate between this and the succeeding latest Blancan assemblage. For tapirs, only Tapirus lundeliusi is known from the middle late Blancan. No records of Tapirus haysii are known from Florida during this interval, but geographic coverage of known faunas is limited, so it may have been living elsewhere in the state, or it could have been temporarily extirpated. The latest Blancan Florida vertebrate interval Sigmodon curtisi Gidley and Sigmodon minor Gidley (replacing S medius of the preceding interval). Other species the latest Blancan are Equus leidyi Hay, Spilogale putorius (Linnaeus) Peromyscus sarmocophinus Ruez and Atopomys texensis Patton Inglis 1A, Inglis 1C, and De Soto Shell Pit are the primary examples of latest Blancan faunas. Both Tapirus lundeliusi and Tapirus haysii are known from this interval. While T. haysii continued into the early Irvingtonian in Florida (Hulbert 1995), there are no younger records of T. lundeliusi SYSTEMATIC PALEONTOLOGY Order PERISSODACTYLA Owen 1848 Family TAPIRIDAE Gray 1821 Genus TAPIRUS Brisson 1762 Subgenus HELICOTAPIRUS subgen. nov. Type Species. Tapirus veroensis Sellards 1918. Included Species. Tapirus haysii Leidy 1859; Tapirus lundeliusi sp. nov. Etymology From the Greek helix (spiral, coil, or curl) plus Tapirus In reference to the spiral shape of the prominent meatal fossa on the frontal and nasal bones of members of this subgenus. Tapirus that share a closer common ancestor with Tapirus veroensis than they do with Tapirus ( Tapirus ) terrestris (Linnaeus 1758), Tapirus ( Pinchacus ) pinchaque (Roulin 1829), Tapirus ( Tapirella ) bairdii (Gill 1865), or Tapirus ( Acrocodia ) indicus Desmarest 1819. Diagnosis.Species of Tapirus that differ from Tapirus ( Tapirus ) and Tapirus ( Pinchacus ) by having anterior process of maxilla visible in lateral view dorsal to premaxilla; large, triangular interparietal; extensive meatal fossa on frontal and nasal whose medial margin lies very close to midline of skull; lacrimal with weak or absent process; relatively shorter postcanine diastema; relatively greater depth of mandibular ramus below molars; anterior margin of ascending ramus projects anteriorly in lateral view; and increased relative length of p2. Additionally they differ from Tapirus ( Tapirus ) by having a much shorter sagittal crest. Additionally they differ from Tapirus ( Pinchacus ) by having shorter nasal and broader P1, and by lacking posterolabial cingula on upper molars. Species of Tapirus that differ from Tapirus ( Tapirella ) by having longer posterior process of cartilage; dorsal surface of nasal and frontal on same plane; long descending nasal process; narrow sagittal crest in adults; posterodorsal process of maxilla widely exposed dorsal to orbit, forming base of trough for meatal diverticulum; relatively

PAGE 15

HULBERT JR.: Blancan Tapirus from Florida 79 shorter postcanine diastema; relatively greater depth of mandibular ramus below molars; anterior margin of ascending ramus projects anteriorly in lateral view; and increased relative length of p2. Species of Tapirus that differ from Tapirus ( Acrocodia ) and Tapirus ( Megatapirus ) by having narrow sagittal crest in adults; dorsal surface of nasal and frontal on same plane; greater exposure of lacrimal in lateral view with at most one visible, large foramen; and premaxillary-maxillary suture extends anterior to alveolar margin of C in lateral view; and by lacking posterolabial cingula on upper molars. Additionally they differ from T. ( Acrocodia ) by having shallower meatal fossa on frontal and nasal with lateral supraorbital wall; no diastemata between incisors; broader P1 with strong lingual cusp; stronger protoloph on P2; fourth metatarsals. Additionally they differ from T. ( Megatapirus on frontal and nasal; relatively shorter braincase; and external auditory meatus not closed ventrally by converging postglenoid and paroccipital processes. Occurrence.Early to latest Pleistocene of the central and eastern United States. Remarks.Each of the four extant species of Tapirus is the type species of a named subgenus: T ( Tapirus ) terrestris ; T ( Pinchacus ) pinchaque ; T ( Tapirella ) bairdii ; and T ( Acrocodia ) indicus (Hershkovitz 1954). Matthew and Granger (1923) named a new subgenus and species of Tapirus T ( Megatapirus ) augustus based on late Pleistocene fossils from China. Starting with Colbert and Hooijer (1953), Megatapirus has generally been used at full generic rank, e.g., McKenna and Bell (1997); Tong et al. (2002), Wu et al. (2006), Louys et al (2007). However, Tong et al. (2002) and Tong (2005) hypothesized that Megatapirus augustus evolved in China from within Tapirus (Colbert & Hooijer 1953:90 also implied this is the case), such that its use at the generic level makes Tapirus paraphyletic. For that reason I treat it here at its original rank, a subgenus of Tapirus Simpson (1945) and Ray and Sanders (1984) admonished against formally naming subgenera for North American fossil species of Tapirus But with the greatly increased number of known fossil species and a better understanding of their cranial morphology and phylogenetic relationships, the clade containing Tapirus veroensis and two other species is clearly distinct from those represented by the four modern subgenera (Hulbert & Wallace 2005; also see phylogeny section below) and deserving of a formal name. Other than those three species, Tapirus polkensis (Olsen) is the most likely to also be included in T ( Helicotapirus ), but its phylogenetic relationships with T ( Tapirella ) and T ( Helicotapirus ) veroensis are equivocal at present (see below), so it is not assigned to the new subgenus. Tapirus webbi Hulbert and Tapirus johnsoni Schultz, Martin, and Corner are clearly not members of T ( Helicotapirus ), and the same is likely also true of Tapirus simpsoni Schultz, Martin, and Corner, Tapirus merriami Frick, and Tapirus californicus Merriam, but they are currently too poorly known to resolve their relationships. All of the relatively well known fossil species of Tapirus from South America (e.g., Tapirus mesopotamicus ), China (e.g., Tapirus hezhengensis Tapirus sanyuanensis ), and Europe (e.g., Tapirus arvernensis Tapirus jeanpiveteaui ) differ from T ( Helicotapirus ) in various combinations of cranial and dental characters (based on descriptions in Boeuf 1991; Rustioni 1992; Tong et al. 2002;Tong 2005; Ferrero & Noriega 2007; Deng et al. 2008), and none appear to be referable to the new subgenus. However, an inclusive phylogenetic analysis will be needed to determine this more securely. Given that the southeastern U.S. was always the primary stronghold of T ( Helicotapirus ), it is logical to look for its origins in Neogene tapirs from this region. In addition to the already mentioned T polkensis other possibilities are the specimens from the late Miocene Withlacoochee River 4A site in Florida described by Hulbert (2005) or the larger, unnamed species of tapir from the early Pliocene Palmetto neither is known from relatively complete skulls, such that their phylogenetic relationships can be determined for the most part only on the basis of small samples of teeth. For tapirids, that rarely produces satisfactory or conclusive results.

PAGE 16

80 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) TAPIRUS ( HELICOTAPIRUS ) LUNDELIUSI SP. NOV. Figures 3 Tapirus sp. Webb 1974, table 2.1 (in part); Webb and Wilkins 1984, table 1. Tapirus n. sp. Morgan and Hulbert 1995, p 68 and table 2; Hulbert 1999, p. 53A. new late Blancan species of Tapirus Hulbert 2005, pp. Holotype.UF 160715, partial young adult skeleton that includes cranium, mandible, vertebrae, ribs, sternebrae, and foreand hind limb elements. Collected 12 December 1994 by A. E. Pratt, R. Toomey, and R. C. Hulbert. Found in near full articulation. Paratype.UF 121736, partial subadult skeleton that includes cranium, mandible, vertebrae, ribs, and foreand hind limb elements. Collected in 1989 and 1990 by R. C. Hulbert and G. S. Morgan. Etymology.The species name honors Ernest L. Lundelius of the University of Texas at Austin for his numerous contributions to the study of Pleistocene mammals, including Tapirus (Lundelius 1972; Lundelius & Slaughter 1976). Type Locality.Haile 7C (see Appendix 1). Occurrence.Very early Pleistocene (late Blancan, ca. 1.6.6 Ma) of peninsular Florida. Possibly also late Blancan of North Carolina. Referred Specimens.Haile 7C: in addition to the holotype and paratype, this locality produced four other skeletons with skulls and mandibles, two young juveniles (UF 162351 and 206877), and two young adults (UF 206876 and 206878), and portions of postcranial skeletons from four other individuals (UF 149299, 160713, 162352, and 177735). Also the following isolated elements: UF 212272, M1; 212273, P3; 212274, I1; 212275, i1; 177736, tibia; 177737, scapula; 177738, patella; 177739, metatarsal 2. Haile 7G: Very young juveniles: UF 224663, 224664, 224668, 224676, 243306, 244089, 244513, 247104. Young juveniles: UF 224666, 224673, 224681, 242904, 244090, 244512, 244514, 244515, 244516, 247101, 247103. Juveniles: UF 223827, 224662, 224665, 244519, 247102. Subadults: UF 224671, 224682, 224684. Young adults: UF 221720, 224672, 224679, 224685, 224686, 244503, 244511. Full adults: UF 224670, 224674, 224680, 224683, 244502, 244504. Old adult: UF 244600. Santa Fe River 8: UF 177509, squamosal; 15095, M2 and M3; 177512, I2; 177513, i3; 177510, axis; 177511, radius; 177514, unciforms; 240325, trapezoid; 177520, scaphoid; 177521, pisiform; 177517, magnums; 177522, cuneiforms; 15096, 177524 metacarpal 3s; 177525, metacarpal 4; 17527 177528, metacarpal 5s; 15098, 177531 cuboids; 240382, navicular; 177536, ectocuneiforms; 177526, metatarsal 2; 177529, metatarsal 3; 177530, metatarsal 4; 15097, 18476, 177539, phalanges. Santa Fe River 8A: UF 177546, frontal; 177547, fused supraoccipital, interparietal, and parietal; 16763, associated(?) P1 and p2; 14261, M3; 177548, thoracic vertebrae; 177717 177718, lunars; 45727, pisiform; 177716, unciform; 177719, metacarpal 2; 177720, metacarpal 4s; 14259, 177715 astragali; 14524, calcaneum; 177712, cuboids; 177714, ectocuneiform; 45728, navicular; 177724, 240225, metatarsal 2s; 14257, metatarsal 3; 177722, metatarsal 4s; 14260, 177725, phalanges. Santa Fe River 8C: UF 177501, scapula; 177502, ulna; 177503, metacarpal 3; 177504, 177507, calcaneum; 177508, metatarsal 4. Withlacoochee River 1A: UF 242909, M3; 241366, metatarsal 4. Waccasassa River 9A: UF 240682, lower molar. Inglis 1A: UF/TRO 2625 maxilla with M1 M3; UF 66134, nasal; 66125, supraoccipital; 18175, maxilla with M1M2; 115970, maxilla with P1P4; 18176, I1; 18177, DP4; 115969, p3; 176712, dp3 or dp4; 176713, two lower cheek tooth fragments; 45305, mandibular symphysis; 22689, 177701, axis vertebrae; 177702, cervical vertebrae; 177704, thoracic vertebrae; 18178, humerus; 45309, 45447, 177835, distal humeri; 177710, lunar; 244473, scaphoids; 244475, magnums; 244478,

PAGE 17

HULBERT JR.: Blancan Tapirus from Florida 81 unciform; 244479, pisiform; 176715, metacarpal 3; 45549, metacarpal 4; 11597115975, metacarpal 5s; 45278, distal femur; 45306, tibiae; 45440, astragalus; 18179, 115978, calcanea; 177711, cuboid; 244480, ectocuneiforms; 22690, fused mesocuneiform, ectocuneiform, and metatarsal 1; 45450, 115977, 176701, metatarsal 2s; 18180, 115976 metatarsal 3s; 45448, metatarsal 4; 18181, 45441, 1767091, phalanges. Inglis 1B: UF 97116, P2; 97117, mandible fragment with m3; 97118, p3; 97119, i2; 97120, metacarpal 3; 97121, cuneiform. Inglis 1D: UF 224247, humerus; 224248, metacarpal 2; 224249, femur; 224250, thoracic vertebra. Walden Pit 2: UF/TRO 1476, mandible fragment with m3. Diagnosis.Species of T ( Helicotapirus ) Tapirus haysii ; smaller on average than Tapirus veroensis but with extensive overlap in most measured characters. Sagittal crest slightly more pronounced than in T. veroensis and T. haysii Interparietal fused to occipital prior to eruption of M2 (fusion in T. veroensis and T. haysii occurs after eruption of M2, sometimes not until after eruption of M3). Only a single large lacrimal foramen present medial to the posterior process of lacrimal; not visible in lateral view ( T. veroensis has two large lacrimal foramina and usually at least one is visible in lateral view). Posterolateral margin of nasal not sharply downturned (present in T. haysii ). Dorsal table of frontal relatively smaller than in T. veroensis and T. haysii with an irregular T. veroensis and T. haysii medial to posterior process of premaxilla (absent in T. veroensis and T. haysii ). P1 lacks a strong transverse loph, but its width is relatively large. TAPIRUS (HELICOTAPIRUS) HAYSII LEIDY 1859 Figure 2 Referred Blancan Specimens.Santa Fe River 1: UF 177841, partial M2 or M3; 177842, associated right and left p4s. Santa Fe River 1B: UF 213920, 224631 M1s (?). Kissimmee River 2: UF 51250, P3. Kissimmee River 6: UF 52601, metacarpal 5. Inglis 1C: UF 177844, P1. De Soto Shell Pit 5: UF 223920, associated right maxilla with DP2-DP4 and left maxilla with DP1-DP4; 232075, cuboid. Devils Elbow 2: UF 177843, M3. See Appendix 1 for locality information. Table 2) are much larger than any found in the Haile 7C and 7G samples of Tapirus lundeliusi and instead fall within or near the OR of Tapirus haysii (= Tapirus copei ) from the Irvingtonian of Florida (Ray 1964; Hulbert 1995) and Pennsylvania (Simpson 1945). They are also similar in size to the holotype and referred specimens of Tapirus haysii from North Carolina (Ray and Sanders 1984). The matching p4s (UF 177842) have the greatest known L and AW for that tooth among all specimens assigned to T. haysii (if western specimens of a large tapir represent a different species, Tapirus merriami as proposed by Jefferson 1989, in contrast to Ray and Sanders 1984). The reference of the Florida Blancan sample to T. haysii is also supported by a transverse loph on the P1, UF 177844 (Fig. 2F), a characteristic feature of this species, and one rarely present in T. lundeliusi or T. veroensis UF 213920 and 224631 are isolated upper molars that are either relatively large M1s or small M2s. As the other Santa Fe River specimens are of relatively large size, notably UF 188742, an assignment as M1s is considered more likely. The dimensions of the metacarpal 5 all fall within the OR of the early Irvingtonian sample of T. haysii from Florida provided by Hulbert (1995), and T. lundeliusi Chronologic Distribution.The range of Tapirus haysii in Florida is emended from that in Hulbert (1995) to be early late Blancan through early Irvingtonian. The lone middle Irvingtonian record of T. haysii in Florida listed by Hulbert (1995) was McLeod Rock Pit, based on two juvenile specimens. Increased numbers of juveniles of Tapirus veroensis no longer support placing the McLeod fossils in T. haysii ; they are instead more likely to be early representatives of T. veroensis

PAGE 18

82 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)

PAGE 19

HULBERT JR.: Blancan Tapirus from Florida 83 Figure 4. Skull of holotype of Tapirus lundeliusi sp. nov., UF 160715, from Haile 7C. A dorsal view; B ventral view; C occlusal view of right P1M3; D right lateral view; and E posterior view (zygomatic arches removed). This skull was restored so that a cast could be made for public display at the Florida Museum of Natural History.

PAGE 20

84 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3)

PAGE 21

HULBERT JR.: Blancan Tapirus from Florida 85 Figure 5. Skull of paratype of Tapirus lundeliusi sp. nov., UF 121736, from Haile 7C. A dorsal view; B ventral view; C occlusal view of muzzle with right and left I1I3, C. D left lateral view (reversed); and E posterior view (zygomatic arches removed). The M2 is not yet fully erupted in this individual.

PAGE 22

86 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) TAPIRUS SUBGENUS AND SPECIES INDETERMINATE Figure 19 Referred Specimens.US 19 Bridge Site: UF 255357, partial juvenile maxilla with DP3 (missing metacone); UF 255358, mandibular fragment with dp3 (missing most enamel on labial side of crown); UF 248788, talonid of p2; UF 247180, navicular; UF 248789, proximal phalanx of digit 2 or 4; Hubbard Pit: UF 249089, dp3; Kissimmee River 6: UF 51249, p2; Sommers Pit: UF/TRO 1469, unworn p4 (crown only, missing roots); Walden Pit 1: UF/TRO 1489, mandibular fragment with posterior root of p2 and moderately worn p3. See Appendix 1 for more information on localities. Measurements.UF 255357: AW, ca. 18.3 mm. UF 255358: L, 19.8 mm. UF 249089: L, 19.5 mm; AW, ca. 12.8 mm; PW, 13.0 mm. UF 51249: L, 20.0 mm; PW, 12.4 mm. UF 248788: PW, 12.4 mm. UF/TRO 1469: L, 20.2 mm; AW, 15.1 mm; PW, 16.7 mm. UF/TRO 1489: L, 17.9 mm; AW, 13.0 mm; PW, 14.2 mm. UF 247180: greatest W of astragalar facet, 29.8 mm; greatest depth of astragalar facet, approximately 21 mm. UF 248789: greatest L, 24.7 mm; proximal W, 16.4 mm; distal W, 13.5 mm. permanent premolar is visible in the crypt medial Table 2. Comparison of dental measurements of late Blancan specimens of Tapirus haysii from Florida with samples from the early Irvingtonian of Florida and the middle Irvingtonian Port Kennedy Cave Site in Pennsylvania (MINMAX (N)). Values for the latter after Simpson (1945). Catalogue Number Tooth Locus Measurement Value (mm) Irvingtonian of Florida Port Kennedy Cave, PA UF 223920 DP4 L 23.6 22.3.1 (5) UF 223920 DP4 AW 26.7 24.9.3 (5) UF 223920 DP4 PW 23.2 22.8.0 (5) UF 177844 P1 L 20.7 19.7.3 (7) 22.4.9 (3) UF 177844 P1 PW 19.6 17.3.0 (7) 19.6.5 (3) UF 51250 P3 L 24.0 21.6.5 (10) 22.7.5 (5) UF 51250 P3 AW 27.5 25.7.0 (10) 27.0.5 (4) UF 51250 P3 PW 26.6 25.4.3 (11) 26.1.0 (4) UF 213920 M1 L 25.5 23.9.9 (12) 25.8.4 (5) UF 213920 M1 AW 31.7 26.9.7 (11) 28.9.1 (5) UF 213920 M1 PW 28.2 24.5.5 (12) 25.8.9 (4) UF 224631 M1 L 26.3 23.9.9 (12) 25.8.4 (5) UF 224631 M1 AW 30.8 26.9.7 (11) 28.9.1 (5) UF 224631 M1 PW 28.0 24.5.5 (12) 25.8.9 (4) UF 177841 M2/M3 AW 32.4 30.8.5 (19) 31.0.9 (9) UF 177843 M3 L 29.9 27.1.1 (6) 26.8.2 (5) UF 177843 M3 AW 32.5 31.5.6 (6) 31.0.1 (5) UF 177843 M3 PW 26.8 25.1.9 (6) 26.5.0 (5) UF 177842 p4 L 28.8 22.5.7 (6) 24.1.9 (6) UF 177842 p4 AW 22.8 19.9.0 (7) 18.3.7 (8) UF 177842 p4 PW 22.7 20.2.0 (6) 19.5.8 (8)

PAGE 23

HULBERT JR.: Blancan Tapirus from Florida 87 based on its relative position to the IOF in similar age individuals, the orientation and height of the protoloph, and strength of the parastyle. UF/TRO ratio of about 90 percent (which eliminates molars) and absence of a metalophid (which eliminates p2 and p3). In contrast, the complete tooth in UF/ TRO 1489 has a metalophid and a small portion of the mental foramen is present ventral to the set of alveoli anterior to the tooth. This combination of features is only present in the p3 of Tapirus on the basis of the presence of metastylids (not found on permanent teeth), relatively narrower but otherwise molariform morphology, and low crown height. They compare very favorably to the dp3 of regarded as possible but less likely because of their relatively low hypsodonty. DESCRIPTION OF TAPIRUS LUNDELIUSI C RANIUM Cranial measurements are listed in Table 3. The rostrum is narrow with a relatively short postcanine diastema (Figs. 4-6). In lateral view, the posterior process of the premaxilla terminates in an acute point approximately dorsal to the anterior margin of the P1; the maxilla is exposed medial to the posterior process of the premaxilla and there forms a slight (UF 160715, 224680, 224682) to modest (UF 121736, 206878, 244502), ca. 30 mm septum (Figs. 4D, 5D, 6-7); and the dorsomedial border of the maxilla thins posterior to the maxillary ventrally). The narial incision terminates dorsal to the middle or posterior half of the orbit (Figs. 4D, 6A). The posterodorsal ascending process of the maxilla forms the margin of the narial aperture back to the middle of the orbit where it articulates with the descending process of the nasal; this portion of the maxilla is widely exposed dorsally, not twisted laterally (Figs. 4A, 5A). The anterior supraorbital process of the frontal overlaps the lacrimal dorsally and articulates medially with the maxilla and descending process of the nasal. Dorsal to the orbit, the frontal forms a broad, concave shelf that carried the meatal diverticulum; posterior to the orbit, this groove on the frontal narrows, curves dorsomedially, and becomes shallower, with a poorly demarcated posterior margin. As the groove approaches the midline of the skull it curves anterolaterally, and becomes deeper and broader as it reaches the nasal. The anterior border of the spiral groove on the nasal (N = 9) and even on the juvenile UF 162351 (Figs. the nasal near its widest point. The shape and size of the preserved nasals vary, but in all subadults and adults the lateral edge is posteriorly convex and anteriorly slightly concave (N = 8) or straight (N = 1), they are slightly arched anteroposteriorly, and thin dorsoventrally (Fig. 6A). Nasal length is equal to or greater than twice maximum nasal width in 8 of 9 subadult and adults. In the one very young juvenile with well preserved nasals (UF 247104, Fig. 9), the meatal fossa is faint and such that there is no osteological contact along the midline of the right and left nasals (there may have been cartilaginous contact). A second very young juvenile, UF 224668, appears to have the same nasal morphology as UF 247104, although both nasals are incomplete such that their lack of contact can not be ascertained unambiguously. Slightly older juveniles, such as UF 162351, have nasals with an overall morphology resembling those of adults, including a long region of contact along the midline. The dorsal table of the frontals (the region bounded anteriorly by the nasals and posterolaterally by the temporal crests) is relatively small because the parasagittal ridges are strongly concave and meet anterior to the frontoparietal suture (Fig. 8). The dorsal surface of the frontals the midline, and a smaller and slightly lower but notably raised region on the parasagittal ridge directly posterior to the descending process of the nasal. Internally, the frontals contain sinuses

PAGE 24

88 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Table 3. Measurements (in mm) on skulls of Tapirus lundeliusi n. sp. from the early Pleistocene of Florida. For comparison, values for other species of Tapirus x s second line is OR(N), and third line is percentage difference of x with mean value of T. lundeliusi ). Occipital HT measured from ventral surface of basioccipital to dorsal surface of supraoccipital. Specimen Number Basal L P1-P4 L M1-M3 L P1-M3 L C-P1 DL Muzzle W Occipital HT Occipital Condyle W UF 160715 330 75.5 69.9 144.3 32.3 43.3 94.5 74.1 UF 121736 334 78.4 43.6 43.4 93.3 71.3 UF 115970 75.45 UF 206876 325 75.2 67.7 142.5 30.0 101.6 78.0 UF 206878 77.6 71.3 147.8 35.2 UF 224672 42.2 UF 224674 78.4 70.3 147.65 36.2 UF 224679 40.3 107.5 81.5 UF 224680 73.1 66.05 138.6 39.9 45.2 79.6 UF 244502 73.6 UF 244503 80.1 72.2 152.3 98.0 86.8 mean s CV 329.7.51 1.37 76.7.27 2.97 70.2.59 3.70 145.5.79 3.29 37.5.82 12.88 44.0.07 2.43 99.0.76 5.82 78.5.49 6.98 T. johnsoni 353(1) 108% 72.7.22 67.2-77.1(4) 95% 66.2.28 63.6-68.6(4) 94% 137.2.44 130.5-143.9(2) 94% 44.4.34 40.4-49.0(3) 118% 43.2(1) 98% 103.0(1) 104% 77.6(1) 99% T. simpsoni 84.9(1) 111% 73.8(1) 105% 159.5(1) 110% 43.9(1) 117% T. polkensis 282.7.74 273-290(3) 86% 60.3.52 52.8-66.6(19) 79% 55.6.43 52.8-62.4(15) 79% 115.6.17 108.2-127.8(9) 79% 33.5.32 26.6-41.9(18) 89% 39.2.76 36.0-41.4(7) 89% 82.9.17 80.6-86.1(6) 84% 65.1.69 63.7-67.4(4) 83% T. haysii 87.2.70 86.7-87.7(2) 114% 80.0.47 77.5-81.3(5) 114% 166.3.41 166.0-166.6(2) 114% 48.5.55 45.6-50.4(3) 129% 52.3.32 48.7-55.9(4) 119% 114.4.92 111.7-117.5(3) 115% 83.0.29 81.5-84.5(4) 106% T. veroensis 353.5.02 345-362(2) 107% 77.9.92 71.8-84.2(12) 102% 70.0.33 63.6-77.9(11) 100% 143.7.92 133.4-159.0(6) 99% 40.7.52 30.0-50.1(11) 108% 48.0.42 45.2-52.5(6) 109% 108.3.43 105.7-112.2(3) 109% 81.7.69 76.1-89.1(3) 104% T. terrestris 350.6.83 328-384(16) 106% 71.8.42 66.0-80.8(18) 94% 65.4.48 61.0-71.4(18) 93% 134.8.93 125.3-147.4(18) 93% 47.9.35 40.7-57.4(17) 128% 43.0.37 37.0-50.0(16) 98% 122.8.99 115.5-132.2(16) 124% 71.4.88 64.7-79.3(16) 91% T. bairdii 379.51.15 361-409(19) 115% 74.9.23 67.2-80.9(21) 98% 65.5.53 61.1-70.8(20) 93% 138.1.14 126.0-148.4(19) 95% 54.0.41 43.5-60.7(22) 144% 50.6.94 44.6-55.1(22) 115% 120.3.12 110.0-132.7(21) 122% 83.4.10 77.1-91.9(19) 106% T. indicus 403.3.37 399-407(3) 122% 89.0.22 84.5-94.7(4) 116% 78.1.52 70.6-82.4(3) 111% 164.01.06 151.8-173.3(3) 113% 53.0.31 40.1-62.4(4) 141% 50.8.14 49.5-52.3(4) 116% 115.7.59 106.5-122.0(4) 117% 88.7.47 86.9-93.9(4) 113%

PAGE 25

HULBERT JR.: Blancan Tapirus from Florida 89 Figure 6. Left lateral views of anterior portions of skulls of Tapirus lundeliusi sp. nov. from Haile 7G. A UF 224682, subadult with C, P1P3, DP4, M1, and erupting M2; B UF 224680, full adult with I2, P1 M3. Abbreviations: asc. proc., ascending process; iof, infraorbital foramen; max., maxilla; ppl, posterior process of lacrimal.

PAGE 26

90 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Figure 7. Left lateral views of anterior portions of skulls of Tapirus lundeliusi sp. nov. from Haile 7G. A UF 247101, very young juvenile with DP1DP4; B UF 224662, juvenile with DP1M1. Abbreviations as in Figure 6.

PAGE 27

HULBERT JR.: Blancan Tapirus from Florida 91 Figure 8. Dorsal views of nasals, frontals, and sagittal crests in adults of Tapirus lundeliusi sp. nov. A UF 160715 from Haile 7C; B UF 224680 from Haile 7G. Note variation in length of nasals.

PAGE 28

92 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) which begin forming in young juveniles. The posteriorly converging parasagittal ridges meet to form the sagittal crest in the posterior portion of the frontal and it continues onto the parietals (Fig. 8). The ontogenetic sequence of skulls from Haile 7C and 7G reveals the developmental pattern of the sagittal crest (Figs. 8-10). Very young juveniles have smoothly rounded parietals without any indication of parasagittal ridges (UF 224664) or very weak ridges far from the midline (UF 224668, of the young juvenile stage (UF 162351), but the ridges thicken and move towards the midline later in this stage, about the same time as the M1 begins to erupt (UF 206877, 244514, 247103). The parasagittal ridges become narrower and taller to form a sagittal crest consisting of two distinct ridges separated by a narrow groove in juveniles and subadults (UF 121736, 224682). In young and full adults (UF 160715, 206876, 206878, 224680), and the crest is slightly taller than in the subadults. crest is slightly arched, reaching an acme only several millimeters higher than the dorsal tables of the frontals or occipital (Figs. 4D, 5D). The parasagittal ridges begin to diverge posterolaterally about 40 to 50 mm from the posterior edge of the triangular-shaped region that is continuous with the dorsal surface of the sagittal crest. Much of this Figure 9. Dorsal view of the braincase in a very young juvenile of Tapirus lundeliusi sp. nov., UF 247104 from Haile 7G.

PAGE 29

HULBERT JR.: Blancan Tapirus from Florida 93 Figure 10. Dorsal views of the skulls of Tapirus lundeliusi sp. nov. A juvenile UF 224662 from Haile 7G; B subadult UF 121736 from Haile 7C. Contrast the development of the sagittal crest and the fusion of interparietal and occipital bones in these individuals to the conditions in the very young juvenile in Figure 9. Scale bars are 5 cm in length.

PAGE 30

94 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) large, triangular interparietal bone (observable in UF 162351, 206877, 224664, 224668, 244514, and 247104). The interparietal fuses with the occipital as early as the young juvenile stage (UF 224666) but more often in the juvenile stage (UF 224662). Fusion of the interparietal with the parietals starts in the subadult stage (UF 224682) and is completed by the young adult stage (UF 160715, 206876). The lambdoidal crests are strongly developed and project primarily posteriorly in young juveniles (Fig. 9), but posterolaterally in subadults and adults (Figs 4A, 10B). In lateral view, the infraorbital foramen is located dorsal to the DP2 in juveniles (Fig. 7; N = 8) and the P3 in subadults and adults (Fig. 6; N = 11); its posterior margin is formed by a very thin (ca. 5mmwide) strut of the maxilla. The lacrimal bone articulates anteriorly with this maxillary strut, dorsally with the ascending process of the maxilla, and ventrally with the jugal (Fig. 7A). Posteriorly, the lacrimal forms the anterior margin of the orbit. The dorsoventral height of the lacrimal greatly exceeds its anteroposterior length, and it has a generally smooth, moderately concave lateral surface (Figs. 5D, 6-7). A distinct anterior lacrimal process is usually absent (N = 8) or merely a small rugosity (N = 2). All specimens in which the bone is preserved have a short, broad, the posterior lacrimal process (and not visible in lateral view; Figs. 6, 11) is a single, large lacrimal foramen (N = 12). On UF 121736, it is 7.5 mm tall and 4 mm wide. There are no other large lacrimal ones located dorsal and lateral to the larger opening (Fig. 11A). Figure 11. Condition of the lacrimal foramen in Tapirus lundeliusi sp. nov. A UF 224673, isolated left lacrimal of young juvenile in posterior view; B UF 224680, left rostrum of adult in posterolateral view, arrow points to lacrimal foramen. Every individual of this species in which the lacrimal is preserved, regardless of ontogenetic age, has a single large lacrimal foramen that is not visible in lateral view.

PAGE 31

HULBERT JR.: Blancan Tapirus from Florida 95 The lateral opening for the external auditory meatus is circular, ca. 15 to 17 mm in diameter; the meatus is partially closed ventrally by a posterior projection of the postglenoid process and an anterior projection of the mastoid process, but the two do not connect (Fig. 4D). An anteroventrally directed ridge branches off the descending lambdoidal crest about halfway through its length and continues down to the end of the mastoid process of the squamosal. This ridge is more pronounced in adults than subadults. The mastoid process of the petrosal is visible wedged between this ridge and the lambdoidal crest. On UF 160715, the mastoid process is oval, about 16 mm wide and 28 mm tall, and faces more posteriorly than it does laterally. On UF 121736, it is slightly narrower and faces posteriorly and laterally in about equal measure. The stylomastoid foramen opens at the ventral suture between the mastoid and squamosal into a shallow groove that runs ventrally down the squamosal just anterior to its suture with the paroccipital process. The long paroccipital process projects much more ventrally than either the postglenoid process or the occipital condyles (Figs. 4D-E, 5D-E). The back of the skull dorsal to the foramen magnum is oriented nearly perfectly vertically. The incisive foramen is long, extending from just posterior to the I3 to the level of P1 (Fig. 4B). In most subadults and young adults the incisive foramen ends lateral to the anterior half of the P1, while in full adults it ends lateral to the posterior half of the P1 or even level with the P1/P2 contact. The hard palate formed by the maxillae and palatines is moderately arched transversely and bears several foramina, the largest of which opens anteriorly into a shallow sulcus at about the same level as the anterior border of the M1. The palate ends along the midline at a point on line with the middle of the M2 with a moderate rugosity in adults (UF 160715, 206876, 206878). The choane (internal nares) are relatively narrow. The medial wall of the orbit is not well preserved in any available specimen. The basicranial region is dominated by a very large lacerate foramen lateral to the basioccipital and posterior to the alisphenoid and postglenoid region of the squamosal (Figs. 4B, 5B). The smaller hypoglossal foramen is located near the posterior margin of the lacerate foramen and is approximately circular with a diameter of 6 mm in adults. The width of the foramen magnum is slightly greater than its height. Minimum ventral width between the occipital condyles in adults is about 12 percent of occipital condyle width (N = 3). M ANDIBLE The ventral border of the ramus is gently curved posterior to the uplifted symphysis (Fig. 12). There are no diastemata between any of the incisors or canine, and the postcanine diastema is relatively short (Table 4). The symphysis has a deeply concave dorsal surface. Posteriorly, margin is thick and well developed (Fig. 13). The posteromedial region of the mandible is concave and in subadults and adults bears pronounced ridges that start at the thickened posterior margin. The large mandibular foramen opens posterodorsally and is located at the same height as the cheekteeth (Fig. 14). The mediolateral axis of the mandibular condyle is oblique, with the medial side lower than the lateral side. Dorsoventral height from the angle to the mandibular condyle is relatively greater than in other species of Tapirus (Table 4). The anterior margin of the ascending ramus projects anteriorly in lateral view (Figs. 12D, 14B) so that at the level of the condyle it partially overlies the m3. The coronoid process is tall, extending 40 mm beyond the height of the condyle. The masseteric fossa is large and deep, with its ventral border crowns of the cheekteeth. The mental foramen is located ventral to the dp2 or p2 (N = 27). One rare variant is UF 244502 with two mental foramina, one ventral to the p2 and another ventral to the p3. This unusual feature is found on both the right and left dentaries of this individual. D ENTITION The incisors and canines are morphologically like those of other New World Tapirus in particular there is a reduced upper canine, a caniniform I3, an enlarged, spatulate i1, and a greatly reduced i3 (Figs. 5C, 12A, C, 14A). The P1 is slightly

PAGE 32

96 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Table 4. Statistics of measurements (in mm) on mandibles of Tapirus lundeliusi sp. nov. from the early Pleistocene of Florida and those of other species of Tapirus x s second line is OR(N). For T. lundeliusi third line is CV ; for other species of Tapirus third line is percentage difference of its mean value with that of T. lundeliusi Condylar L p2-p4 L m1-m3 L p2-m3 L c-p2 DL Muzzle W H at p2 H at m3 condyle H T. lundeliusi 283.9.70 278-290(2) 3.06 63.6.06 59.9-66.1(8) 3.24 72.6.98 69.1-77.6(8) 4.03 135.5.82 129.9-143.3(6) 3.56 45.5.53 38.5-49.1(11) 7.75 46.2.79 43.5-50.6(6) 6.03 42.0.18 39.2-44.6(9) 5.20 61.1.19 56.2-66.7(8) 5.22 151.3.34 139.7-162.9(5) 6.84 T. polkensis 243.2.76 237-255(5) 86% 50.0.33 46.8-53.8(16) 79% 57.5.83 52.4-61.6(16) 79% 107.2.14 101.2-114.4(11) 79% 39.2.37 32.5-47.0(18) 86% 36.9.24 32.5-39.2(10) 80% 33.1.63 27.4-37.7(13) 79% 44.8.10 38.3-49.8(14) 73% 112.4.08 109.0-116.5(6) 74% T. johnsoni 298.0.54 288-307(3) 105% 59.7.44 55.8-63.0(13) 94% 69.4.48 66.0-73.6(12) 96% 128.6.41 122.0-136.0(11) 95% 58.1.20 52.5-564.2(4) 128% 42.7.14 40.4-44.6(3) 92% 46.1.00 42.4-49.7(4) 110% 56.8.98 53.9-60.8(9) 93% 136.9.34 134.7-140.0(4) 90% T. webbi 329.3.89 326.5-332(2) 116% 65.6.87 62.1-68.3(8) 103% 76.6.42 72.2-79.4(6) 105% 143.1.14 137.5-147.5(4) 106% 58.1.96 55.2-61.2(6) 128% 50.0.21 46.6-55.7(4) 108% 44.8.92 41.1-46.6(6) 107% 56.3.52 50.0-61.7(11) 92% 128(1) 85% T. haysii 343.6.54 326-359(4) 121% 72.9.50 67.7-78.5(6) 115% 82.8.83 76.7-89.6(6) 116% 158.1.13 153.4-166.1(5) 117% 59.4.11 52.4-68.4(5) 131% 52.9.60 51.1-54.2(3) 114% 52.6.15 45.5-58.2(7) 125% 71.6.15 67.3-78.2(5) 118% 173.0.65 171-176(3) 114% T. veroensis 308(1) 109% 63.9.06 59.1-67.1(6) 100% 72.0.02 65.9-76.8(22) 99% 134.2.39 133.9-134.7(3) 99% 47.4.08 41.9-53.4(7) 104% 44.7.28 41.1-48.0(6) 97% 44.3.91 42.1-49.8(6) 105% 59.6.44 53.7-65.9(23) 97% 131.1.20 128.2-136.2(4) 87% T. terrestris 286.7.31 267-315(16) 101% 60.4.16 55.9-67.1(16) 95% 67.6.78 63.0-74.8(16) 93% 127.2.76 118.5-141.1(15) 94% 55.5.35 48.2-67.6(17) 122% 44.2.04 36.6-53.8(15) 96% 45.5.29 37.6-55.2(17) 108% 51.9.64 46.9-57.4(16) 85% 123.3.76 111.2-133(12) 82% T. bairdii 304.0.60 289-323(21) 107% 62.0.56 58.3-67.9(19) 97% 66.8.33 61.7-70.9(20) 92% 128.2.74 119.7-138.5(18) 95% 65.3.61 53.9-73.1(22) 144% 51.1.46 47.6-61.4(22) 111% 47.6.15 41.7-51.3(22) 113% 55.2.09 51.3-64.6(21) 90% 133.5.73 123.5-144(15) 88% T. indicus 336.8.88 332.5-340(3) 119% 74.4.70 69.5-77.6(4) 117% 81.1.43 70.4-87.3(4) 112% 155.2.31 137.9-165.3(4) 115% 58.3.99 53.3-64.3(4) 128% 55.8.97 50.8-64.2(4) 121% 54.1.46 51.6-56.5(3) 129% 60.4.15 58.1-62.3(3) 99% 155.5(1) 103%

PAGE 33

HULBERT JR.: Blancan Tapirus from Florida 97 Table 4. Statistics of measurements (in mm) on mandibles of Tapirus lundeliusi sp. nov. from the early Pleistocene of Florida and those of other species of Tapirus x s second line is OR(N). For T. lundeliusi third line is CV ; for other species of Tapirus third line is percentage difference of its mean value with that of T. lundeliusi Condylar L p2-p4 L m1-m3 L p2-m3 L c-p2 DL Muzzle W H at p2 H at m3 condyle H T. lundeliusi 283.9.70 278-290(2) 3.06 63.6.06 59.9-66.1(8) 3.24 72.6.98 69.1-77.6(8) 4.03 135.5.82 129.9-143.3(6) 3.56 45.5.53 38.5-49.1(11) 7.75 46.2.79 43.5-50.6(6) 6.03 42.0.18 39.2-44.6(9) 5.20 61.1.19 56.2-66.7(8) 5.22 151.3.34 139.7-162.9(5) 6.84 T. polkensis 243.2.76 237-255(5) 86% 50.0.33 46.8-53.8(16) 79% 57.5.83 52.4-61.6(16) 79% 107.2.14 101.2-114.4(11) 79% 39.2.37 32.5-47.0(18) 86% 36.9.24 32.5-39.2(10) 80% 33.1.63 27.4-37.7(13) 79% 44.8.10 38.3-49.8(14) 73% 112.4.08 109.0-116.5(6) 74% T. johnsoni 298.0.54 288-307(3) 105% 59.7.44 55.8-63.0(13) 94% 69.4.48 66.0-73.6(12) 96% 128.6.41 122.0-136.0(11) 95% 58.1.20 52.5-564.2(4) 128% 42.7.14 40.4-44.6(3) 92% 46.1.00 42.4-49.7(4) 110% 56.8.98 53.9-60.8(9) 93% 136.9.34 134.7-140.0(4) 90% T. webbi 329.3.89 326.5-332(2) 116% 65.6.87 62.1-68.3(8) 103% 76.6.42 72.2-79.4(6) 105% 143.1.14 137.5-147.5(4) 106% 58.1.96 55.2-61.2(6) 128% 50.0.21 46.6-55.7(4) 108% 44.8.92 41.1-46.6(6) 107% 56.3.52 50.0-61.7(11) 92% 128(1) 85% T. haysii 343.6.54 326-359(4) 121% 72.9.50 67.7-78.5(6) 115% 82.8.83 76.7-89.6(6) 116% 158.1.13 153.4-166.1(5) 117% 59.4.11 52.4-68.4(5) 131% 52.9.60 51.1-54.2(3) 114% 52.6.15 45.5-58.2(7) 125% 71.6.15 67.3-78.2(5) 118% 173.0.65 171-176(3) 114% T. veroensis 308(1) 109% 63.9.06 59.1-67.1(6) 100% 72.0.02 65.9-76.8(22) 99% 134.2.39 133.9-134.7(3) 99% 47.4.08 41.9-53.4(7) 104% 44.7.28 41.1-48.0(6) 97% 44.3.91 42.1-49.8(6) 105% 59.6.44 53.7-65.9(23) 97% 131.1.20 128.2-136.2(4) 87% T. terrestris 286.7.31 267-315(16) 101% 60.4.16 55.9-67.1(16) 95% 67.6.78 63.0-74.8(16) 93% 127.2.76 118.5-141.1(15) 94% 55.5.35 48.2-67.6(17) 122% 44.2.04 36.6-53.8(15) 96% 45.5.29 37.6-55.2(17) 108% 51.9.64 46.9-57.4(16) 85% 123.3.76 111.2-133(12) 82% T. bairdii 304.0.60 289-323(21) 107% 62.0.56 58.3-67.9(19) 97% 66.8.33 61.7-70.9(20) 92% 128.2.74 119.7-138.5(18) 95% 65.3.61 53.9-73.1(22) 144% 51.1.46 47.6-61.4(22) 111% 47.6.15 41.7-51.3(22) 113% 55.2.09 51.3-64.6(21) 90% 133.5.73 123.5-144(15) 88% T. indicus 336.8.88 332.5-340(3) 119% 74.4.70 69.5-77.6(4) 117% 81.1.43 70.4-87.3(4) 112% 155.2.31 137.9-165.3(4) 115% 58.3.99 53.3-64.3(4) 128% 55.8.97 50.8-64.2(4) 121% 54.1.46 51.6-56.5(3) 129% 60.4.15 58.1-62.3(3) 99% 155.5(1) 103% Figure 12. Mandibles of Tapirus lundeliusi sp. nov. from Haile 7C. A & C occlusal views of symphysis; B & D lateral views of left mandible. AB UF 121736, subadult paratype with i1, c, p2p3, dp4, m1m2; CD UF 160715, young adult holotype with i3, c, p2m3. Scale bar of 5 cm on right applies to both B and D.

PAGE 34

98 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) longer than wide (Figs. 4C, 15A-C, 16C; Table 5); its morphology is best shown on the slightly worn UF 16763, 121736, 224671, and 224682. The labial portion of the P1 crown is united in a sharp ectoloph; the paracone and metacone are situated close to each other; and the paracone is the tallest cusp on the P1 although neither it nor the metacone extend much beyond the level of the remainder of the ectoloph (Fig. 6A). A narrow, low, posterior cingulum wraps around the back of the P1, connecting the ectoloph with the bulbous, main lingual cusp. A slight crest extends labially from the lingual cusp towards the metacone in four of seven unworn or slightly worn P1s, while the other three lack any trace of a transverse loph. About half of the P1s have a much smaller and lower cuspule located anterior to the main lingual cusp, while the other half lack such a structure. With wear, the P1 ectoloph remains a sharp ridge (UF 160715, 115920, 206876, 224686, 224674), while the cuspule (if present). The AW of the P2 is on average 91 percent of PW (N = 18). The P2 metacone and paracone Figure 13. Posterior view of right and left mandibles of UF 160715, holotype of Tapirus lundeliusi sp. nov.

PAGE 35

HULBERT JR.: Blancan Tapirus from Florida 99 Figure 14. Mandibles of Tapirus lundeliusi sp. nov. A occlusal view of symphysis, and B medial view of right mandible (reversed) of UF 206878, adult mandible with i1i3, c, p2m3 from Haile 7C; C medial view of UF 221720, young adult mandible with right c and left p2m3 from Haile 7G. Arrow in B points to location of mandibular foramen. 5 cm scale bar applies to B and C.

PAGE 36

100 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) are well separated, unlike those of the P1, but in the unworn state also barely project beyond the level of the rest of the ectoloph (Figs 6A, 15E, 16C). The P2 has a small parastyle and anterior and posterior cingula. The transverse lophs are oriented obliquely, especially the weak protoloph that extends from the protocone towards the parastyle where it connects near the base of the ectoloph. The metaloph is oriented more transversely than the protoloph, and is taller, uniting about halfway along the height of the ectoloph. As in the succeeding teeth, the metaloph curves posteriorly just prior to uniting with the metaloph, so that it connects with the metacone. The P3 (Figs. 4C, 15A-C, 16C; Table 5) is more molariform than the P2 in that AW exceeds PW in most individuals (14 of 16) and the transverse lophs are not as oblique. Compared to the P2, the protoloph is stronger and unites with the paracone instead of the parastyle, and the parastyle is larger. When unworn, the P3 ectoloph is sharply depressed between the paracone and metacone. The P4 continues these trends, basically being a larger version of the P3 with a stronger parastyle and taller cusps and lophs (Figs. 4C, 15A-C). The M1 through M3 also have large parastyles that are separated from the paracones by deep creases; AW greatly exceeds PW, especially on the M3. A small cusp is variably present on the anterior side of the protoloph close to the anterior cingulum. Small lingual and labial cuspules are also variably present between the protocone and hypocone, or paracone and metacone, respectively, of the P3M3. None of the teeth have a cingulum around the labial base of the metacone. The DP1 is approximately the same size as the P1 (Table 7), slightly lower crowned, and has a well formed but narrow transverse loph the extends from the lingual cusp to the ectoloph (Fig. 16A). The DP2 is about the same length as the P2, but much narrower, and has a stronger protoloph, a much lower crown, and an equally weak parastyle. The DP3 and especially the DP4 are morphologically more like lower crowned molars than their permanent successors, with strong styles and PW shorter than AW (Fig. 16). As in the permanent premolars, no posterolabial cingula are present on the deciduous premolars. Lower cheekteeth (Figs. 12B, D, 14C, 17; Table 6) have the classic bilophodont morphology of Tapirus Hypolophids on premolars are complete and in unworn teeth are only slightly lower than the protolophids (Figs. 12B, 17A). Protolophids and hypolophids on unworn molars are of equal height. The p2 has a well developed metalophid (cristid obliqua) and paralophid; the former blocks the transverse valley between the protolophid and hypolophid. Average p2 L/p3 L is 112 percent (N = 12; OR = 10419). The p3 also has a pronounced metalophid, but it is not as tall as that of the p2. L of the p3 is similar to that of the p4 (Table 6), but the p3 is on average narrower, such that mean PW/L is 80 percent in p3 and 87 percent in p4 (N = 14). AW is always less than PW in the p3, with a mean AW/PW of 91 percent. AW is slightly greater relative to PW in the p4, with a mean AW/PW of 94 percent, and AW slightly exceeds PW in 2 of 13 individuals (UF 244502 and 224672). The p4m3 lack a metalophid. Prominent anterior and posterior cingulids are present on unworn and slightly worn p3m3, but lateral and lingual cingulids are absent. The anterior and posterior cingulids are soon obliterated by interdental wear (Fig. 17). The posterior cingulid of the m3 lacks a prominent hypoconulid, and instead has only a slightly thickened central stylid. With rare exceptions (e.g., UF 206876, 206878), AW on m1 and m2 exceeds PW, with very similar mean AW/PW ratios, 104 percent in m1 (N = 16) and 103 percent in m2 (N = 12). The PW of the m3 is relatively narrower than that of the m1 and m2, with a mean AW/PW of 111 percent (N = 12). The lower deciduous premolars are on average longer, narrower, and less high crowned than their permanent replacements (Fig. 18; Table 7). The dp2 has a strong metalophid like the p2, but the dp3 does not. The hypolophids are complete and as tall as the protolophids on the dp2dp4. Small metastylids are present in unworn and slightly worn dp2dp4 on the posterolingual side of the metaconids; these are rapidly lost to wear. The dp3 and dp4 are molariform in their overall

PAGE 37

HULBERT JR.: Blancan Tapirus from Florida 101 Figure 15. Adult upper cheek teeth of Tapirus lundeliusi sp. nov. in occlusal view. A UF 224674, left P1M3, from Haile 7G; B UF 224680, left P1M3, from Haile 7G; C UF 115970, left P1P4, from Inglis 1A; D UF 18175, right M1M2 (reversed), from Inglis 1A; E UF 97116, right P2 (reversed), from Inglis 1B; F UF 242909, partial right M3 (reversed) from Withlacoochee River 1A.

PAGE 38

102 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Table 5. Measurements of individual cheek teeth of the holotype, paratype, and referred specimens of Tapirus lundeliusi sp. nov. from the Haile 7C locality, Alachua County, Florida. Locus L AW PW Locus L AW PW UF 160715 P1 18.5 17.4 UF 160715 P2 19.1 21.3 22.4 p2 21.9 14.9 UF 160715 P3 19.4 23.7 23.0 p3 19.1 14.4 16.0 UF 160715 P4 20.8 25.6 24.5 p4 21.0 16.8 18.5 UF 160715 M1 21.9 26.3 22.5 m1 21.9 17.0 16.8 UF 160715 M2 24.4 28.0 23.3 m2 24.1 18.1 17.3 UF 160715 M3 25.3 28.2 22.6 m3 24.6 18.3 17.2 UF 121736 P1 18.1 16.2 UF 121736 P2 19.4 20.6 22.2 p2 23.8 14.4 UF 121736 P3 19.9 24.1 22.7 p3 21.0 14.3 16.0 UF 121736 DP4 20.9 24.8 21.5 dp4 22.2 15.5 15.6 UF 121736 M1 22.1 26.0 22.9 m1 22.9 17.5 16.8 UF 121736 M2 23.7 25.8 22.2 m2 23.7 UF 121736 M3 25.6 26.5 22.1 m3 25.8 18.1 17.5 UF 206876 P1 17.2 15.2 UF 206876 P2 19.8 21.2 22.9 p2 23.4 14.6 UF 206876 P3 20.2 23.8 23.4 p3 20.8 15.4 17.0 UF 206876 P4 21.3 24.3 23.8 p4 21.4 17.1 18.8 UF 206876 M1 22.0 25.8 23.1 m1 22.5 17.4 17.8 UF 206876 M2 23.7 27.0 23.5 m2 24.0 18.0 18.2 UF 206876 M3 24.7 26.8 22.1 m3 26.0 18.3 16.4 UF 206878 P1 19.3 15.8 UF 206878 P2 19.3 21.8 24.1 p2 22.7 15.4 UF 206878 P3 20.1 25.3 24.5 p3 21.5 15.2 17.2 UF 206878 P4 21.2 26.6 25.5 p4 21.7 17.4 19.7 UF 206878 M1 21.6 26.1 22.5 m1 23.3 17.5 18.1 UF 206878 M2 24.5 27.6 24.0 m2 25.4 18.8 19.2 UF 206878 M3 25.3 27.8 22.3 m3 26.2 18.6 18.0 UF 212272 M1 22.4 26.7 23.8

PAGE 39

HULBERT JR.: Blancan Tapirus from Florida 103 Figure 16. Juvenile and subadult upper cheek teeth of Tapirus lundeliusi sp. nov. from Haile 7C and 7G in occlusal view. A UF 224673, right DP1DP4; B UF 224662, right DP1M1; C UF 121736, right P1P3, DP4, M1M2. Upper scale bar applies to A and B; lower scale bar to C.

PAGE 40

104 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Table 6. Univariate statistics for permanent cheekteeth of Tapirus lundeliusi sp. nov. from the early Pleistocene of Florida and Tapirus veroensis from the middle and late Pleistocene of Florida, Georgia, and South Carolina. Rightmost two columns show the results of twotailed t p Tapirus lundeliusi sp nov. Tapirus veroensis N x s MIN MAX CV N x s MIN MAX CV t P1 L 16 18.56 0.86 17.0 19.8 4.65 38 18.77 1.06 15.7 20.6 5.67 -0.699 NS P1 PW 16 17.00 1.00 15.2 18.4 5.88 38 16.44 1.53 13.7 19.4 9.31 1.334 NS P2 L 18 19.44 0.84 18.0 21.6 4.34 39 19.84 0.97 18.0 21.9 4.90 -1.513 NS P2 AW 18 21.45 0.90 19.7 22.9 4.21 39 21.31 1.04 18.9 23.3 4.89 0.479 NS P2 PW 18 23.51 0.87 22.2 25.1 3.71 35 23.56 1.23 21.4 25.8 5.21 -0.134 NS P3 L 18 19.97 0.55 18.6 20.7 2.74 36 20.49 0.94 18.9 22.6 4.60 -2.166 p <0.05 P3 AW 15 24.64 0.77 23.7 26.1 3.14 35 24.74 1.08 22.4 26.6 4.35 -0.322 NS P3 PW 16 24.08 0.89 22.7 25.7 3.68 33 24.69 1.16 22.3 26.7 4.70 -1.839 NS P4 L 17 21.35 0.60 19.9 22.2 2.81 48 21.65 1.25 18.7 23.6 5.76 -0.963 NS P4 AW 15 26.60 0.87 24.3 27.5 3.26 46 26.93 1.37 24.0 29.3 5.10 -0.872 NS P4 PW 16 25.33 0.81 23.8 26.9 3.20 47 26.27 1.40 23.0 28.9 5.33 -2.553 p <0.05 M1 L 18 22.35 0.76 20.7 23.9 3.41 51 22.78 1.16 20.2 25.5 5.09 -1.456 NS M1 AW 18 26.56 0.56 25.8 27.7 2.12 47 26.48 1.16 24.1 28.8 4.39 0.287 NS M1 PW 17 23.41 0.62 22.5 24.9 2.67 48 24.07 1.02 22.2 26.3 4.24 -2.520 p <0.05 M2 L 18 24.57 0.96 22.7 26.2 3.89 55 24.93 1.16 22.8 27.6 4.67 -1.185 NS M2 AW 18 28.29 1.02 25.8 29.6 3.60 52 28.75 1.42 25.0 31.9 4.94 -1.256 NS M2 PW 16 24.33 1.01 22.2 26.0 4.17 50 25.80 1.17 23.1 28.3 4.53 -4.516 p <0.001 M3 L 16 25.23 0.85 23.4 27.0 3.38 56 24.80 1.24 22.0 27.9 5.02 1.283 NS M3 AW 15 27.84 0.96 26.2 29.2 3.45 50 28.48 1.65 24.7 31.7 5.79 -1.422 NS M3 PW 15 22.86 1.17 21.1 25.8 5.11 50 24.07 1.44 20.3 27.0 5.98 -2.971 p <0.05

PAGE 41

HULBERT JR.: Blancan Tapirus from Florida 105 Table 6. Univariate statistics for permanent cheekteeth of Tapirus lundeliusi sp. nov. from the early Pleistocene of Florida and Tapirus veroensis from the middle and late Pleistocene of Florida, Georgia, and South Carolina. Rightmost two columns show the results of twotailed t p Tapirus lundeliusi sp nov. Tapirus veroensis N x s MIN MAX CV N x s MIN MAX CV t P1 L 16 18.56 0.86 17.0 19.8 4.65 38 18.77 1.06 15.7 20.6 5.67 -0.699 NS P1 PW 16 17.00 1.00 15.2 18.4 5.88 38 16.44 1.53 13.7 19.4 9.31 1.334 NS P2 L 18 19.44 0.84 18.0 21.6 4.34 39 19.84 0.97 18.0 21.9 4.90 -1.513 NS P2 AW 18 21.45 0.90 19.7 22.9 4.21 39 21.31 1.04 18.9 23.3 4.89 0.479 NS P2 PW 18 23.51 0.87 22.2 25.1 3.71 35 23.56 1.23 21.4 25.8 5.21 -0.134 NS P3 L 18 19.97 0.55 18.6 20.7 2.74 36 20.49 0.94 18.9 22.6 4.60 -2.166 p <0.05 P3 AW 15 24.64 0.77 23.7 26.1 3.14 35 24.74 1.08 22.4 26.6 4.35 -0.322 NS P3 PW 16 24.08 0.89 22.7 25.7 3.68 33 24.69 1.16 22.3 26.7 4.70 -1.839 NS P4 L 17 21.35 0.60 19.9 22.2 2.81 48 21.65 1.25 18.7 23.6 5.76 -0.963 NS P4 AW 15 26.60 0.87 24.3 27.5 3.26 46 26.93 1.37 24.0 29.3 5.10 -0.872 NS P4 PW 16 25.33 0.81 23.8 26.9 3.20 47 26.27 1.40 23.0 28.9 5.33 -2.553 p <0.05 M1 L 18 22.35 0.76 20.7 23.9 3.41 51 22.78 1.16 20.2 25.5 5.09 -1.456 NS M1 AW 18 26.56 0.56 25.8 27.7 2.12 47 26.48 1.16 24.1 28.8 4.39 0.287 NS M1 PW 17 23.41 0.62 22.5 24.9 2.67 48 24.07 1.02 22.2 26.3 4.24 -2.520 p <0.05 M2 L 18 24.57 0.96 22.7 26.2 3.89 55 24.93 1.16 22.8 27.6 4.67 -1.185 NS M2 AW 18 28.29 1.02 25.8 29.6 3.60 52 28.75 1.42 25.0 31.9 4.94 -1.256 NS M2 PW 16 24.33 1.01 22.2 26.0 4.17 50 25.80 1.17 23.1 28.3 4.53 -4.516 p <0.001 M3 L 16 25.23 0.85 23.4 27.0 3.38 56 24.80 1.24 22.0 27.9 5.02 1.283 NS M3 AW 15 27.84 0.96 26.2 29.2 3.45 50 28.48 1.65 24.7 31.7 5.79 -1.422 NS M3 PW 15 22.86 1.17 21.1 25.8 5.11 50 24.07 1.44 20.3 27.0 5.98 -2.971 p <0.05 Table 6. Continued. Tapirus lundeliusi sp nov. Tapirus veroensis N x s MIN MAX CV N x s MIN MAX CV t p2 L 15 23.55 0.89 21.9 24.7 3.79 40 23.62 1.12 20.9 25.5 4.75 -0.232 NS p2 PW 14 14.71 0.94 12.8 15.9 6.40 38 14.88 1.18 11.8 17.3 7.91 -0.474 NS p3 L 15 21.00 1.10 19.1 23.2 5.23 47 21.39 1.21 18.0 24.6 5.65 -1.127 NS p3 AW 14 15.33 0.66 14.3 16.4 4.31 48 15.97 1.18 14.1 19.0 7.40 -1.953 NS p3 PW 14 16.80 0.57 16.0 17.7 3.38 48 18.02 1.37 15.5 20.9 7.57 -3.259 p <0.01 p4 L 14 21.67 1.04 19.8 23.6 4.82 48 22.24 1.29 19.6 24.5 5.78 -1.519 NS p4 AW 13 17.65 0.70 16.4 19.3 3.96 47 18.94 1.27 15.2 21.2 6.73 -3.496 p <0.01 p4 PW 14 18.78 0.68 17.5 19.9 3.64 46 20.27 1.46 17.4 23.4 7.22 -3.678 p <0.01 m1 L 17 23.13 0.99 21.5 25.2 4.26 50 22.55 1.07 20.3 25.8 4.77 1.956 p <0.05 m1 AW 17 18.16 0.82 17.0 20.0 4.54 48 18.71 1.01 16.6 20.8 5.40 -2.011 p <0.05 m1 PW 16 17.35 0.67 16.1 18.5 3.84 49 17.56 0.95 15.8 20.6 5.40 -0.801 NS m2 L 13 25.15 1.23 23.4 26.8 4.89 66 25.45 1.23 21.8 27.6 4.83 -0.811 NS m2 AW 13 19.27 0.91 18.0 21.7 4.70 65 20.18 0.97 17.8 22.3 4.80 -3.113 p <0.01 m2 PW 12 18.61 0.91 17.3 20.5 4.88 65 19.21 1.03 16.0 21.8 5.35 -1.914 NS m3 L 13 26.32 1.20 24.6 29.3 4.56 63 26.63 1.38 23.9 29.8 5.19 -0.737 NS m3 AW 12 19.23 1.10 18.1 21.9 5.72 62 19.95 1.00 18.0 22.8 5.01 -2.247 p <0.05 m3 AW 13 17.32 0.70 16.4 18.8 4.02 63 17.90 0.95 16.2 20.3 5.33 -2.070 p <0.05

PAGE 42

106 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Figure 17. Lower cheek teeth of Tapirus lundeliusi sp. nov. from Haile 7C and 7G in occlusal view. A UF 121736 with right p2p3, dp4, m1, and erupting m2; B UF 221720 with left p2m2 and erupting m3 (reversed); C UF 160715 with right p2m3; D UF 207868 with right p2m3; E UF 224680 with right p2m3.

PAGE 43

HULBERT JR.: Blancan Tapirus from Florida 107 Figure 18. Left mandible of very young juvenile of Tapirus lundeliusi sp. nov., UF 224664, from Haile 7G in A lateral; B occlusal; and C medial views. Teeth present are fully erupted but essentially unworn dp2 dp3, fully formed but unerupted dp4 in crypt, and forming m1 in crypt (partially visible through broken area just anterior to mandibular foramen in C). Upper scale bar applies to A and C, lower scale bar to B.

PAGE 44

108 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Table 7. Univariate statistics for deciduous cheekteeth of Tapirus lundeliusi sp. nov. from the early Pleistocene of Florida and Tapirus veroensis from the middle and late Pleistocene of Florida, Georgia, and South Carolina. Rightmost two columns show the results of twotailed t p Tapirus lundeliusi n. sp. Tapirus veroensis N x s MIN MAX CV N x s MIN MAX CV t DP1 L 14 18.17 0.82 16.6 19.4 4.54 5 17.71 0.85 16.4 18.5 4.82 1.065 NS DP1 PW 14 16.91 0.60 16.0 18.0 3.55 5 16.48 1.01 15.4 17.6 6.13 1.136 NS DP2 L 14 19.86 0.60 18.4 20.9 3.04 14 20.00 1.03 18.0 21.3 5.15 -0.447 NS DP2 AW 9 18.95 0.60 18.0 19.9 3.17 13 18.59 0.98 16.5 20.2 5.28 0.985 NS DP2 PW 12 20.11 1.07 18.2 21.4 5.33 13 19.96 0.94 18.6 21.2 4.69 0.393 NS DP3 L 12 20.73 0.71 19.4 22.0 3.42 12 20.84 0.95 19.5 22.4 4.56 -0.334 NS DP3 AW 11 21.31 0.77 20.3 22.8 3.61 12 21.32 0.74 19.7 22.3 3.48 -0.030 NS DP3 PW 10 20.97 1.10 19.2 22.2 5.23 10 20.82 0.67 19.3 21.7 3.20 0.370 NS DP4 L 10 21.56 0.71 20.8 22.8 3.28 11 22.43 1.03 20.6 23.5 4.58 -2.256 p <0.05 DP4 AW 10 24.26 0.44 23.3 24.8 1.81 11 25.22 0.88 23.8 26.9 3.51 -3.097 p <0.01 DP4 PW 8 21.76 0.86 20.6 23.1 3.96 11 22.53 1.29 20.2 24.0 5.73 -1.477 NS dp2 L 12 25.67 0.72 24.5 27.0 2.79 9 25.92 1.19 24.2 27.7 4.59 -0.599 NS dp2 PW 10 14.59 0.64 13.6 15.6 4.36 10 15.06 0.80 13.8 16.0 5.28 -1.441 NS dp3 L 16 21.62 0.66 20.6 22.7 3.04 8 22.26 1.15 20.8 24.1 5.16 -1.758 NS dp3 AW 13 14.78 0.74 13.8 16.1 5.02 7 15.66 0.92 14.3 16.7 5.87 -2.336 p <0.05 dp3 PW 16 14.93 0.79 13.6 16.3 5.28 8 15.68 1.11 14.1 17.0 7.07 -1.918 NS dp4 L 13 23.11 0.61 22.2 24.3 2.65 19 23.32 1.03 21.8 24.7 4.42 -0.665 NS dp4 AW 11 16.07 0.34 15.5 16.5 2.14 19 17.36 0.86 16.1 19.2 4.94 -4.743 p <0.001 dp4 PW 12 16.19 0.62 15.3 17.3 3.82 20 16.64 0.69 15.6 18.0 4.13 -1.852 NS

PAGE 45

HULBERT JR.: Blancan Tapirus from Florida 109 proportions and morphology, although differ from true molars in that mean AW and PW are about the same (Figs. 17A, 18B). Isolated dp4s and m1s can be distinguished on the basis of AW/L ratios; those with ratios less than 73 percent are dp4s, greater than 73 percent, m1s. Lower crown height and thinner enamel also characterize the dp4 relative to the m1. DESCRIPTION AND COMPARISONS OF TAPIRUS SP. FROM SOMMERS PIT, WALDEN PIT 1, KISSIMMEE RIVER 6, HUBBRD PIT, AND US 19 BRIDGE SITE A species of Tapirus smaller in average size localities in Florida. The isolated p4 from Sommers Pit (UF/TRO 1469) is unworn and probably had not fully erupted prior to death (Fig. 19D). The hypolophid is very slightly lower in height than the protolophid. The tooth is smaller than any p4 referred to Tapirus lundeliusi Its L does fall within the OR of T. lundeliusi but that is misleading as the anterior and posterior cingulids have not been removed by interdental wear with the adjacent teeth. Its L is only matched by those of T. lundeliusi with well worn teeth, such as UF 224680. The AW of UF/TRO 1469 (15.1 mm) is about 3.5 standard deviations less than the mean of T. lundeliusi while its PW (16.7 mm) is smaller by about 3 standard deviations. Both widths are below the minimum values recorded for T. lundeliusi and Tapirus veroensis (Table 6). The p3 of UF/TRO 1489 (Fig. 19C) from Walden Pit 1 is from a smaller, older individual than UF/TRO 1469. Its L, AW, and PW are all smaller than the minimum values of p3s of T. lundeliusi and T. veroensis (Table 6), often by a large margin. UF 51249 is a moderately worn p2, with the characteristic paralophid and metalophid of tapirid p2s (Fig. 19B). Its L and PW are smaller than the minimum values of p2s of T. lundeliusi and the minimum p2 L for T. veroensis (Table 6). The narrowest p2s of T. veroensis have widths similar to UF 51249, but such specimens are rare. The PW of UF 248788 is identical to that of UF 51249; L can not be measured due to breakage. UF 249089 (Fig. 19E-F) from Hubbard Pit is a slightly worn dp3 (or, less likely, a dp4) that is missing the roots; minor damage to the anterolingual base of the crown prevents accurate determination of its AW. Both the L and PW of UF 249089 fall below the OR of either the dp3 or the dp4 of T lundeliusi and T. veroensis (Table 7). The ratio of PW/L in UF 249089 is 0.67. The dp3s of T. lundeliusi have similar ratios ( x = 0.685, N = 15, OR = 0.64.74), while the dp4s are slightly broader ( x = 0.701, N = 12, OR = 0.68.73). Most (22 of 27) dp3s and dp4s of T. veroensis are also The navicular (UF 247180) has the characteristic shape of this element in Tapirus but is of small size. Greatest transverse width of its astragalar facet, 29.8 mm, is 4.0 standard deviations below the mean for a sample of naviculars of T. lundeliusi and well below its observed range ( x = 35.7; s = 1.49; OR = 31.9.1; N = 22). Slight breakage on UF 247180 prevents accurate measurement of the depth of the astragalar facet, but the value would have been very close to 21 mm. Again, this value lies outside the range for T. lundeliusi ( x = 25.5; s = 1.25; OR = 23.5.7; N = 22). In contrast, the dimensions of UF 247180 are very similar to those of the Gray Fossil Site sample of Tapirus polkensis (astragalar facet width: x = 29.2; s = 1.96; OR = 26.1.3; N = 17; astragalar facet depth: x = 21.3; s = 1.14; OR = 20.0.9; N = 17). body size than any extant species, are known from the Hemphillian (late Mioceneearly Pliocene) of Florida and Tennessee (Fig. 19A; Hulbert 2005; Hulbert et al. 2009), but have not previously been reported from the Pleistocene of North America. The p3 of UF/TRO 1489 is larger than average members of the Tennessee sample of Tapirus polkensis but falls within that populations OR (Hulbert et al. 2009). When compared to the two species of Tapirus in the late Hemphillian Palmetto Fauna of Florida (Webb et al. 2008), the L and AW of UF/TRO 1489 more closely resemble those of the larger, unnamed species, while its PW is more similar to that of T. polkensis In contrast, UF/TRO

PAGE 46

110 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) 1469 is larger than p4s referred to T. polkensis from both Florida and Tennessee (Hulbert et al. 2009), and its L, AW, and PW are all within the ORs of the larger, unnamed Palmetto Fauna tapir. The L of the p2, UF 51249, is greater than any referred specimen of T. polkensis from Florida or Tennessee (n = 31), and instead is very similar to the two known p2s of the larger, unnamed species in the Palmetto Fauna (19.3 for UF 23948 and 20.1 for UF 124191). Its PW (along with that of UF 248788), however, is slightly less than those of these two specimens, and within the range of T. polkensis The L and PW of UF 249089 and L of UF 255358 exceed those of dp3s of T. polkensis from both Florida Figure 19. Lower premolars of Tapirus sp. indet. from Florida in occlusal ( A-E ) and lingual ( F ) views. A UF 124191, left partial mandible with p2-p4 from Whidden Creek Site, Fort Meade Mine, Polk County, Florida, very late Hemphillian; B UF 51249, left p2 from Kissimmee River 6; C UF/TRO 1489 right p3 (reversed) from Walden Pit 1; D UF/TRO 1469 right p4 (reversed) from Sommers Pit; E occlusal and F lingual, views of UF 249089, left dp3 from Hubbard Pit. The Blancan teeth in B-F are too small to be referred to Tapirus lundeliusi or Tapirus haysii but are instead within the size range of an undescribed, small species from the late Hemphillian of Florida. Upper scale bar applies to A-D; lower scale bar to E-F.

PAGE 47

HULBERT JR.: Blancan Tapirus from Florida 111 and Tennessee (Hulbert et al. 2009), and are most similar to those of the larger, unnamed Palmetto Fauna tapir, especially UF 223991 (L = 19.9; AW = 13.0; PW = 13.2). Given the close proximity of Sommers Pit and Walden Pit 1 and their equivalent geologic age, it is reasonable to assume that UF/TRO 1469 and UF/TRO 1489 represent the same species. It is statistically improbable that they represent Tapirus lundeliusi (recall that UF/TRO 1476 from nearby Walden Pit 2 falls within the OR of T. lundeliusi ). Of the two species of Tapirus known from the late Hemphillian of Florida, overall they are more similar in size to the unnamed, relatively larger species than they are to Tapirus polkensis (Fig. 19). The same applies to UF 249089. The p2 from Kissimmee River 6 and the teeth from the US 19 Bridge Site show similarities to both late Hemphillian species, but they too more favor the unnamed, relatively larger species rather than T. polkensis The two postcranial elements from the the same bones in T. lundeliusi Other than size, no preserved morphologic features favor assignment to one of these Hemphillian species over the other. Another alternative is that they represent a completely different species, one not present in the Hemphillian, as equivalence in tooth size is not proof that two samples of Tapirus At the present all of these hypotheses are viable. Whatever the case, this small tapir, added to Tapirus haysii and T. lundeliusi brings the total number of species of Tapirus in Florida during the early late Blancan to three. PHYLOGENETIC RELATIONSHIPS Figures 20 show the results of cladistic analyses to investigate the evolutionary relationships between Tapirus lundeliusi sp. nov. and other New World species of Tapirus (Appendix 3). The analyses differ in the presence/absence of the recently described late Miocene Chinese species Tapirus hezhengensis Deng et al. 2008. Given that it is only known from three relatively incomplete specimens, most characters for T. hezhengensis must be scored as missing. But the relative antiquity of this species, Figure 20. Cladograms of the two equally most parsimonious arrangements of North American fossil species and extant Neotropical species of Tapirus ( A and B ), and the strict consensus tree ( C ). Multi-state characters treated as unordered; state matrix in Appendices 2 and 3. Taxon abbreviations: Out, outgroup; Tjo, T johnsoni ; Twe, T webbi ; Tpi, T pinchaque ; Tte, T terrestris ; Tpo, T polkensis ; Tba, T. bairdii ; Tlu, T lundeliusi ; Tha, T haysii ; and Tve, T veroensis along with its geographic position intermediate between North America and Europe suggest that it is likely to be phylogenetically informative with regards to tapir evolution. Recovery of additional specimens of T. hezhengensis to allow scoring of additional characters would likely help resolve inconsistencies in the analyses.

PAGE 48

112 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) In a cladistic analysis using all of the taxa in Appendix 3 with the exception of Tapirus hezhengensis ordered and unordered multi-state characters produced the same relationships among in-group taxa (Fig. 20). With unordered characters, there are two most parsimonious cladograms, each with an overall consistency index of 0.68 and retention index of 0.74. The analyses strongly support a monophyletic clade comprised of T. lundeliusi Tapirus haysii and Tapirus veroensis (i.e., the subgenus Helicotapirus ), and that T lundeliusi is the sister group to T haysii plus T veroensis The interrelationships of Tapirus bairdii and Tapirus polkensis with Helicotapirus are unresolved, but there is strong support that they are more closely related to Helicotapirus than to other species of Tapirus Synapomorphies of T ( Helicotapirus ) are concave lateral surface of lacrimal (character 30 in Appendix 2); anterior projection of ascending ramus of mandible (39); increased depth of horizontal ramus below molars (40); enlarged parastyles on upper molars (55); loss of posterolabial cingula on upper cheekteeth (56); and increased relative length of the p2 (69). Synapomorphies of T haysii and T veroensis relative to T lundeliusi are delayed fusion of interparietal with occipital (9); relatively short increased length of lacrimal (29); taller metaloph on P2 (51); and reduced p3 cristid obliqua (73). The clade consisting of Helicotapirus T bairdii and T polkensis is supported by seven unambiguous synapomorphies: broad dorsal table of frontal (4, reversed in T lundeliusi ); triangular interparietal (8); meatal fossa on nasal deep (14) and approaches midline of skull (15); meatal fossa on frontal extensive with a distinct posterior margin (16, reversed in T lundeliusi posterior lacrimal process (33); and relatively short postcanine diastema (37, reversed in T. bairdii ). The trichotomy among Helicotapirus T bairdii and T polkensis is not resolved in this analysis. To date no North American fossil species of Tapirus has been described that shares some of the autapomorphic character states of T. bairdii such as presence of a broad sagittal table instead of a narrow sagittal crest, nasal notch extending posterior to orbit, loss of descending sigmoid process of nasal bone, of nasal septum, shortened posterior process of premaxilla, very long postcanine diastema, and small cheektooth dimensions relative to body size (and skull length). The distribution of cranial and dental characters of such a fossil species could resolve the trichotomy. Addition of Tapirus hezhengensis to the analysis does not effect support for the following monophyletic clades: Tapirus lundeliusi Tapirus haysii and Tapirus veroensis (and their relative positions); Tapirus ( Helicotapirus ), Tapirus bairdii and Tapirus polkensis ; and Tapirus terrestris plus Tapirus pinchaque (Fig. 21). Two differences between T. ( Helicotapirus ), T. bairdii and T. polkensis is resolved, albeit weakly (Bremer support of 1), with T. bairdii sharing a closer relationship with T. ( Helicotapirus ) than does T. polkensis Second, the possibility of a close relationship between Tapirus webbi from the late Miocene Florida with the two extant South American species, (1995; then using the name Tapirus simpsoni for the samples later named T webbi ), is revived (Fig. 21A). A late Miocene dispersal of T webbi or a closely related form into South America, followed by in situ evolution that produced T. terrestris and T. pinchaque is congruent with molecular evidence of a relative ancient (pre-Pliocene) divergence time between T. bairdii and the clade formed by T. terrestris and T. pinchaque (Ashley et al. 1996; Norman & Ashley 2000). More recently, Rincn et al. (2009) noted the resemblance between some fragmentary tapirid material from the late Pliocene or early Pleistocene of Venezuela and T webbi Although their geologic and chronologic evidence remains controversial, Campbell and co-authors have persistently proposed a late Miocene dispersal to South America of a few North American ungulates, including a tapir (Campbell et al. 2000, dispersal of ground sloths from South America to North America (e.g., Morgan 2005). A much later

PAGE 49

HULBERT JR.: Blancan Tapirus from Florida 113 (Pleistocene) dispersal of tapirs into South America appears less likely, as no known North American Pliocene or Pleistocene species is closely related to T. terrestris and T. pinchaque This scenario would have to involve an as yet undiscovered taxon or clade of tapirids that persisted isolated in Central America from the late Miocene to the Pleistocene. DISCUSSION Nineteenth century paleontologists noted that they could not distinguish the smaller fossil teeth of Pleistocene Tapirus found in North America from that of the extant lowland tapir of South America (e.g., Carpenter 1846; Leidy 1859, 1889; Cope 1871), and, in one of the rare instances of taxonomic conservatism on their part, failed to name it as a new species. It took the discovery of a skull to provide convincing evidence that the North American form was a distinct species, Tapirus veroensis differing in many ways from each of the modern species of tapir (Sellards 1918; Simpson 1945; Lundelius & Slaughter 1976). Likewise, if the Blancan remains here named Tapirus lundeliusi consisted of only jaw fragments, isolated teeth, and postcranial elements, then they could not be easily differentiated from T veroensis (Tables 3, 4, 6, 7). Fortunately, that is not the case, as the combined sample of skeletons from the Haile 7C and 7G sites make it one of the best known fossil species of tapir, second only to Tapirus polkensis (Hulbert et al. 2009). One of the interesting observations of the large sample of skulls of Tapirus polkensis from the Gray Fossil Site in Tennessee was the unexpected Figure 21. Cladograms of the three equally most parsimonious arrangements of North American fossil species and extant Neotropical species of Tapirus (abbreviations as in Fig. 20) and the Miocene species Tapirus hezhengensis (The) from China ( A B and C ), and the strict consensus tree ( D ). Multi-state characters treated as unordered; Bremer support values.

PAGE 50

114 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) features that show greater consistency in extant species and Tapirus veroensis for example in the form of the sagittal crest and number of lacrimal foramina (Hulbert et al. 2009). Such is not the case with the Haile 7G sample of Tapirus lundeliusi which is almost as numerous as the Gray Site sample; once ontogenetic variation is factored, is similar to that observed in extant species of Tapirus Characters that present a high degree of variation in the Haile 7G population, e.g. shape of the nasal bone and development of the lingual cusp and transverse loph on the P1, are also variable in large samples of all species of Tapirus However, many skulls and mandibles from Haile 7G remain to be prepared, so this observation must remain tentative for now. One of the unusual aspects of the Haile 7C and 7G sample of Tapirus lundeliusi is the rarity of old adult individuals. Only one out of 44 individuals that can be aged is in this age class (UF 244600), and it only has very small areas of exposed dentine on its m3, so it barely falls into this category. Tapirs from the old adult wear stage are not uncommon at other Florida fossil localities (pers. obs.) or at the Gray Fossil Site (Hulbert et al. 2009). Indeed, the lone tooth from the Withlacoochee River 1A sample of T lundeliusi is at the old adult stage as is UF/TRO 1476 from Walden Pit 2. It is as yet unclear whether the taphonomic factors causing the very large sample of tapir skeletons at the two Haile 7 sites were acting independent of age (catastrophically, although not likely a single, rapid event), or were actively biased against older, more experienced individuals. This will probably not be resolved until the cause of the mortality of the Haile tapirs is better understood. Although there is broad overlap in all tooth dimensions, Tapirus lundeliusi different from Tapirus veroensis in 13 out of 37 tooth measurements in the permanent dentition (Table 6) and four out of 19 in the deciduous dentition (Table 7). One consistent dissimilarity is smaller in T lundeliusi The other difference is the and p4 (AW and PW) in T lundeliusi That the m1 of T lundeliusi T. veroensis is likely an artifact of the demographics of the two measured samples. The sample of T lundeliusi includes a much greater proportion of juveniles and subadults with unworn and slightly worn m1s which have not lost their anterior and posterior cingulids, while the T veroensis sample contains a greater proportion of full and old adults in which the m1 has worn against the adjacent p4 and m2, reducing its measurable length. Including only unworn or slightly worn m1s from the T veroensis sample produces a population with a mean L of 22.9 that of T. lundeliusi The greatest differences in the of DP4 and dp4 in T lundeliusi (Table 7). In their discussion of the type locality of Tapirus haysii Ray and Sanders (1984) noted that the USNM collection contained specimens of both T haysii and Tapirus veroensis from the Neuse River near New Bern, North Carolina. (Ray & Sanders 1984:293), but resembles mixed BlancanRancholabrean faunas in Florida, such as Santa Fe River 1 (compare with MacFadden & Hulbert 2009:table 1). For example the presence of Nannippus and T. haysii are evidence of a Blancan component, while Mammuthus Bison and Castoroides together indicate a Rancholabrean age. Thus the specimens Ray and Sanders (1984) T. veroensis may simply be standard Rancholabrean specimens of that taxon. However, given that there is a now a second species of Tapirus of similar size as T. veroensis from the Blancan, then the smaller Neuse River tapir specimens need to be critically compared with Tapirus lundeliusi Although currently known from a limited geographic range in Florida, it is likely that future discoveries of late Blancan faunas elsewhere in the southeastern United States will produce specimens of Tapirus lundeliusi Fossil species of Tapirus tend to have rather broad geographic distributions, as do three of the four extant species. Even if its geographic range is currently small, it has broad implications

PAGE 51

HULBERT JR.: Blancan Tapirus from Florida 115 for the phylogeny of Tapirus The mixture of plesiomorphic character states in T. lundeliusi (e.g., nasal, early fusion of interparietal) together with apomorphies shared with Tapirus veroensis and Tapirus haysii (e.g. short postcanine diastema, deep horizontal ramus, absence of posterolabial cingula on upper cheekteeth) allows a better sense of the evolutionary relationships and character state transformations between Miocene species such as Tapirus johnsoni and Tapirus polkensis and the later Pleistocene species. Figure 22. Mounted skeleton of Tapirus lundeliusi sp. nov. on public display at the Florida Museum of Natural History. Postcranial skeleton belongs to UF 206878 (with minor restoration). Skull and mandibles (excluding symphysis and incisors) are casts of UF 160715 (with restoration); mandibular symphysis and incisors are a cast of UF 206878 (see Fig. 14A). Skeletal pose designed by Gina Gould; skeleton mounted and casts made and painted by Steve and Sue Hutchens; photography by Jeff Gage.

PAGE 52

116 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) CONCLUSIONS Tapirs are a common component of the large mammalian herbivore guild in Florida during the late Blancan (early Pleistocene). Two named species of Tapirus are recognized from this interval, the very large Tapirus haysii and the medium-sized Tapirus lundeliusi sp. nov. (Fig. 22), plus a smaller, unnamed form making a total of three. But there is only possible one case when more than one species has been found at the same locality (Kissimmee River 6 with T. haysii and the small, unnamed species), suggesting some type of ecologic separation or competitive exclusion was the norm. As the Kissimmee Rive r specimens were collected from spoil piles created by dredging, it is not certain that the two taxa are in fact contemporaneous at this locality. Tapirus lundeliusi is the best represented late Blancan tapir from Florida, recognized from eleven sites, although complete specimens derive from just two, Haile 7C (the type locality) and Haile 7G. Tapirus haysii is rarer and known from only a few Blancan sites in Florida (Fig. 1; Appendix 1). T. haysii became more common and widespread in Florida during the early Irvingtonian (Hulbert 1995). In contrast to T. haysii no Irvingtonian records of T. lundeliusi are known, with the species apparently becoming extinct near the end of the Blancan. The third late Blancan species, known from Walden Pit 1, Sommers Pit, Hubbard Pit, Kissimmee River 6, than T. lundeliusi and may represent a holdover of an undescribed species from the late Hemphillian of Florida. Almost all standard measurements on the skull, mandible, and teeth show considerable overlap between Tapirus lundeliusi and the common middle to late Pleistocene species Tapirus veroensis But T. lundeliusi retains a number of plesiomorphic features in its skull that are more derived in T. veroensis (and Tapirus haysii ), such Formation of the sagittal crest and fusion of the interparietal bone with the occipital and parietals occur earlier in ontogeny in T. lundeliusi compared to T. veroensis (using tooth eruption sequences to age specimens). T. lundeliusi shares with T. haysii a much greater relative mandibular condyle height than found in T. veroensis or other species of Tapirus Phylogenetic analysis using 39 characters provides strong support for a North American clade composed of Tapirus lundeliusi Tapirus veroensis and Tapirus haysii T. lundeliusi is the sister taxon to the other two. This clade is formally recognized as a new subgenus, Helicotapirus Phylogenetic analysis also strongly supports a clade comprised of Helicotapirus the late Mioceneearly Pliocene species Tapirus polkensis and the extant species Tapirus bairdii Many of the synapomorphies of this clade are osteological features of the nasal and frontal bones related to the development of the meatal diverticulum, a soft tissue structure of uncertain function (Witmer et al. 1999). The Helicotapirus T. bairdii T. polkensis clade most likely diverged from that containing the other two extant New World species ( Tapirus terrestris and T. pinchaque ) in the late Miocene. ACKNOWLEDGMENTS This is University of Florida Contribution to Paleobiology number 624. Any project spanning two decades has received assistance and encouragement in many ways. Early research and travel to collections to obtain comparative measurements were undertaken during 1994 while the author was on the faculty of Georgia Southern University and funded by the Georgia Southern University Foundation and Faculty Research Subcommittee. Excavations at Haile 7C in 1994 were funded by NSF grant EAR to S. D. Emslie and those at Haile 7G in 2006 were funded by NSF grant EAR to J. Bloch and RCH. Permission to collect fossils at Haile 7C and 7G was provided by L. Rogers and Limestone Products, Inc. Many hundreds of public volunteers worked with dozens of UF personnel and students at Haile 7C and 7G to collect a vast number of fossil skeletons. Special praise goes to G. Morgan and A. Poyer for their extensive work with RCH at Haile 7C in 1989

PAGE 53

HULBERT JR.: Blancan Tapirus from Florida 117 crew at Haile 7C (S. Emslie, A. Pratt, B. and R. Toomey, and S. and S. Hutchens); R. Portell for his discovery of the Haile 7G locality; and my primary Bourque, A. Hastings, and E. Simons. J. Waldrop discovered and donated specimens from Sommers and Walden Pits. S. and S. Hutchens discovered and donated specimens from the Inglis 1D and Waccasassa River 9A sites. A. Gipson discovered and donated specimens from the Withlacoochee River 1A site. The Kissimmee River, US 19 Bridge Site, and Hubbard Pit specimens were donated by H. Converse, A. Kerner, and J. Smith, respectively. Other recent donors of tapir fossils to UF include S. Alter, M. Anderson, J. Bryan, R. Dykeman, L. Geary, C. Jeremiah, H. Means, M. Renz, P. Roth, M. and S. Searle, D. Thulman, B. and J. Toomey, and the family of the late R. Piotrowski. I thank M. Colbert, C. Ray, E. Scott, B. Shockey, and S. Wallace for useful discussions over the years on tapirid morphology and systematics. Helpful comments and suggestions for improvement were provided by J. Miller who served as editor of this issue and by two outside reviewers. And I especially wish to thank E. Lundelius for his guidance and mentoring during the beginning phase of my career. LITERATURE CITED Ashley, M. V., J. E. Norman, & L. Stross. 1996. Phylogenetic analysis of the perissodactylan family Tapiridae using mitochondrial cytochrome c oxidase (COII) sequences. Journal of Mammalian Evolution 3(4):315. Bader, R. S. 1957. Two Pleistocene mammalian faunas from Alachua County, Florida. Bulletin of the Florida State Museum, Biological Sciences 2(5):53. Bell, C. J., E. L. Lundelius Jr., A. D. Barnosky, R. W. Graham, E. H. Lindsay Jr., D. R. Ruez, Jr., H. A. Semken Jr., S. D. Webb, & R. K. Zakrzewski. 2004. The Blancan, Irvingtonian, and Rancholabrean mammal ages. Pp. 232 314 in M. O. Woodburne, ed. Late Cretaceous and Cenozoic Mammals of North America. Columbia University Press, New York. Boeuf, O. 1991. Tapirus jeanpiveteaui nov. sp., nouvelle espce de Tapiridae (Perissodactyla) du Pliocne de Charente (France). Palaeontographica, Abteilung A 217:177. Brisson, A. D. 1762. Regnum Animale in Classes IX distributum sive synopsis methodica. Edito altero auctior. Theodorum Haak, Leiden, Netherlands, 294 pp. Campbell Jr., K. E., C. D. Frailey, & L. RomeroPittman. 2000. The late Miocene gomphothere Amahuacatherium peruvium (Proboscidea: Gomphotheriidae) from Amazonian Peru: implications for the Great American Faunal Interchange. Instituto Geolgico Minero y Metalrgico, Serie D: Estudios Regionales, Boletn 23:1. Campbell Jr., K. E., C. D. Frailey, & L. RomeroPittman. 2006. The Pan-Amazonian Ucayali Peneplain, late Neogene sedimentation in Amazonia, and the birth of the modern Amazon River system. Palaeogeography, Palaeoclimatology, Palaeoecology 239:166 219. Campbell Jr., K. E., D. R. Prothero, L. RomeroPittman, F. Hertel, & N. Rivera. 2010. Amazonian magnetostratigraphy: dating Interchange. Journal of South American Earth Sciences 29:619. Carpenter, W. M. 1846. Remarks on some fossil bones brought to New Orleans from Tennessee and Texas. American Journal of Science, second series 1:244. Colbert, M. W. 2005. The facial skeleton of the early Oligocene Colodon (Perissodactyla, Tapiroidea). Paleontologica Electronica 8:12A. Retrieved from http://palaeoelectronica.org/2005_1/colbert12/issue1_05. html Cope, E. D. 1871. Preliminary report on the Vertebrata discovered in the Port Kennedy bone cave. Proceedings of the American Philosophical Society 12:73. Dalquest, W. W. & Schultz, G. E. 1992. Ice Age Mammals of Northwestern Texas. Midwestern State University Press, Wichita Falls, Texas, 309 p.

PAGE 54

118 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) Deng, T., W. He, & S. Q. Chen. 2008. A new species of late Miocene tapir (Perissodactyla, Tapiridae) from the Linxia Basin in Gansu, China. Vertebrata PalAsiatica 46(3):190. [in Chinese with English summary] DeSantis, L. G. & MacFadden, B. J. 2007. Identifying forested environments in Deep Time using fossil tapirs: evidence from evolutionary morphology and stable isotopes. Courier Forschung-Institut Senkenberg 258:147. DeSantis, L. R. G. & MacFadden, B. J. 2008. Stable isotopes and REEs of ungulates and xenarthrans from Haile 7G: elucidating the paleoecology of a late Pliocene fossil site from Florida. Journal of Vertebrate Paleontology 28(3, supplement):71A. Desmarest, A. G. 1819. Tapirs fossils. Pp. 458-461. In Nouveau Dictionnaire dHistoire Naturelle, Nouvelle Edition. Chez Deterville, Paris, vol. 32. Emslie, S. D. 1992. Two new late Blancan avifaunas from Florida and extinctions of wetland birds in the Plio-Pleistocene. Natural History Museum of Los Angeles County Science Series 36:249. Emslie, S. D. 1998. Avian community, climate, and sea-level changes in the Plio-Pleistocene of the Florida Peninsula. Ornithological Monographs, No. 50, 113 p. Ferrero, B. S., & J. I. Noriega. 2007. A new upper Pleistocene tapir from Argentina: remarks Neotropical Tapiridae. Journal of Vertebrate Paleontology 27(2):5041. Gibbard, P. L., M. J. Head, M. J. C. Walker, and the Subcommission on Quaternary Stratigraphy. System/Period and the Pleistocene Series/ Epoch with a base at 2.58 Ma. Journal of Quaternary Science 25(2):96-102. Gill, T. N. 1865. [Description of the genus Elasmognathus ]. Proceedings of the Academy of Natural Sciences Philadelphia 17:183. Graham, R. W. 2003. Pleistocene tapir from Hill Top Cave, Trigg County, Kentucky and a review of Plio-Pleistocene tapirs of North America and their paleoecology. Pp. 8718 in B. Schubert, J. I. Mead, & R. W. Graham, eds. Vertebrate Paleontology of Caves. Indiana University Press, Bloomington. Gray, J. E. 1821. On the natural arrangements of vertebrose animals. London Medical Repository Review, 15:296. Hershkovitz, P. 1954. Mammals of northern Colombia, preliminary report no. 7: tapirs (genus Tapirus ), with a systematic review of American species. Proceedings of the United States National Museum 103:465. Hulbert Jr., R. C. 1988. A new Cormohipparion (Mammalia, Equidae) from the Pliocene (latest Hemphillian and Blancan) of Florida. Journal of Vertebrate Paleontology 7:451 468. Hulbert Jr., R. C. 1995. The giant tapir, Tapirus haysii from Leisey Shell Pit 1A and other Florida Irvingtonian localities. Bulletin of the Florida Museum of Natural History 37:515 551. Hulbert Jr., R. C. 1999. Nine million years of Tapirus (Mammalia, Perissodactyla) from Florida. Journal of Vertebrate Paleontology 19(3, supplement):53A. Hulbert Jr., R. C. 2005. Late Miocene Tapirus (Mammalia, Perissodactyla) from Florida, with description of a new species, Tapirus webbi Bulletin of the Florida Museum of Natural History 45(4):465. Hulbert Jr., R. C., J. I. Bloch, & A. R. Poyer. 2006. Exceptional preservation of vertebrates from Haile 7G, a new late Pliocene site from Florida. Journal of Vertebrate Paleontology 26(3, supplement):78AA. Hulbert Jr., R. C., & G. S. Morgan. 1993. Quantitative and qualitative evolution in the giant armadillo Holmesina (Edentata: Pampatheriidae) in Florida. Pp. 134 in R. A. Martin & A. D. Barnosky, eds. Morphologic Change in Quaternary Mammals of North America. Cambridge University Press, Cambridge. Hulbert Jr., R. C., & G. S. Morgan. 2009. Two

PAGE 55

HULBERT JR.: Blancan Tapirus from Florida 119 new late Blancan vertebrate sites from Florida, with a reassessment of late Pliocene vertebrate biochronology in Florida. Abstract Program, Second Annual Meeting of the Southeastern Association of Vertebrate Paleontology, Virginia Museum of Natural History, Martinsville. Retrieved from http:// web.me.com/dooleyclan/SEAVP/SEAVP_ Hulbert Jr., R. C. & S. C W allace 2005. Phylogenetic analysis of late Cenozoic Tapirus (Mammalia, Perissodactyla). Journal of Vertebrate Paleontology, 25(3, supplement):72A. Hulbert Jr., R. C., S. C. Wallace, W. E. Klippel, & P. W. Parmalee. 2009. Cranial morphology and systematics of an extraordinary sample of the late Neogene dwarf tapir, Tapirus polkensis (Olsen). Journal of Paleontology 83(2):238 262. Jefferson, G. T. 1989. Late Cenozoic tapirs (Mammalia: Perissodactyla) of western North America. Natural History Museum of Los Angeles County Contributions in Science 406:1. Jones, D. S., B. J. MacFadden, S. D. Webb, P. A. Mueller, D. A. Hodell, & T. M. Cronin. 1991. Integrated geochronology of a classic Pliocene fossil site in Florida: Linking marine and terrestrial biochronologies. The Journal of Geology 99(5):637. Kurtn, B., & Anderson, E. 1980. Pleistocene Mammals of North America. Columbia University Press, New York, 442 p. L eidy, J. 1859. Descriptions of vertebrate fossils. Pp. 99 in F. S. Holmes, ed. Post-Pleiocene Fossils of South Carolina. Russell and Jones, Charleston. Leidy, J. 1889. Description of vertebrate remains from Peace Creek, Florida. Transactions of the Wagner Free Institute of Science of Philadelphia 2:19. Louys, W., D. Curnoe, & H. W. Tong. 2007. Characteristics of Pleistocene megafauna extinctions in Southeast Asia. Palaeogeography, Palaeoclimatology, Palaeoecology 243:152. Lundelius, E. L. 1972. Fossil vertebrates from the late Pleistocene Ingleside Fauna, San Patricio County, Texas. Report of Investigation, Bureau of Economic Geology, University of Texas, No. 77, 74 pp. Lundelius, E. L., & B. H. Slaughter. 1976. Notes on American Pleistocene tapirs. Pp. 226 243 in C. S. Churcher, ed. Athlon: Essays in Paleobiology in Honour of Loris Shano Russell. Royal Ontario Museum, Toronto. MacFadden, B. J., & R. C. Hulbert. 2009. Calibration of mammoth ( Mammuthus ) dispersal into North America using rare earth elements of Plio-Pleistocene mammals from Florida. Quaternary Research 71(1):41. Matthew, W. D., & W. Granger. 1923. New fossil mammals from the Pliocene of Szechuan, China. Bulletin of the American Museum of Natural History 48:563-598. of Mammals above the Species Level. Columbia University Press, New York, 631 p. Merriam, J. C. 1913. Tapir remains from late University of California Publications, Bulletin of the Department of Geology 7:169. Meylan, P. A. 1982. The squamate reptiles of the Inglis 1A fauna (Irvingtonian: Citrus County, Florida). Bulletin of the Florida State Museum 27(1):1. Morgan, G. S. 2005. The Great American Biotic Interchange in Florida. Bulletin of the Florida Museum of Natural History 45(4):2711. Morgan, G. S. 2008. Vertebrate fauna and geochronology of the Great American Biotic Interchange in North America. Pp. 93 in S. G. Lucas, G. S. Morgan, J. A. Spielmann, & D. R. Prothero, eds. Neogene Mammals. New Mexico Museum of Natural History & Science Bulletin 44. Morgan, G. S., & S. D. Emslie. 2010. Tropical from the Pliocene and Pleistocene of Florida. Quaternary International 217:143. Morgan, G. S., & R. C. Hulbert Jr. 1995. Overview

PAGE 56

120 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) of the geology and vertebrate biochronology of the Leisey Shell Pit local fauna, Hillsborough County, Florida. Bulletin of the Florida Museum of Natural History 37:1. Morgan, G. S., & J. A. White. 1995. Small mammals (Insectivora, Lagomorpha, and Rodentia) from the early Pleistocene (early Irvingtonian) Leisey Shell Pit local fauna, Hillsborough County, Florida. Bulletin of the Florida Museum of Natural History 37:397 462. Norman, J. E., & M. V. Ashley. 2000. Phylogenetics of Perissodactyla and tests of the molecular clock. Journal of Molecular Evolution 50(1):11. Olsen, S. J. 1960. Age and faunal relationships of Tapiravus remains from Florida. Journal of Paleontology 34:164. Owen, R. 1848. Description of teeth and portions of jaws of two extinct anthracotheroid quadrupeds ( Hyopotamus vectianus and Hyop. bovinus ) discovered by the Marchioness of Hastings in the Eocene deposits on the N.W. coast of the Isle of Wight: with an attempt to of pachyderms by the number of their toes. Quarterly Journal of the Geological Society of London 4:103. Petuch, E. J. 1982. Notes on the molluscan paleoecology of the Pinecrest Beds at Sarasota, Florida, with the description of Pyruella a stratigraphically important new genus (Gastropoda, Melongenidae). Proceedings of the Academy of Natural Sciences, Philadelphia 134:12. Ray, C. E. 1964. Tapirus copei in the Pleistocene of Florida. Quarterly Journal of the Florida Academy of Sciences 27:59. Ray, C. E. & A. E. Sanders. 1984. Pleistocene tapirs in the eastern United States. Pp. 283 315 in H. H. Genoways & M. R. Dawson, eds. Contributions in Quaternary Vertebrate Paleontology: a Volume in Memorial to John E. Guilday. Carnegie Museum of Natural History, Pittsburgh, Special Publication 8. Ray, C. E., E. Anderson, & S. D. Webb. 1981. The Blancan carnivore Trigonictis (Mammalia: Mustelidae) in the eastern United States. Brimleyana 5:1. Rincn, A. D., G. E. Parra, F. J. Prevosti, M. T. Alberdi, and C. J. Bell. 2009. A preliminary assessment of the mammalian fauna from the Pliocene-Pleistocene el Breal de Orocual locality, Monagas State, Venezuela. Pp. 593 620 in L. B. Albright III, ed. Papers on Geology, Vertebrate Paleontology, and Biostratigraphy in Honor of Michael O. Woodburne. Museum of Northern Arizona Bulletin 65. Robertson, J. S. 1976. Latest Pliocene mammals from Haile XV A, Alachua County, Florida. Bulletin of the Florida State Museum 20:111 186. Roulin, X. 1829. Memoir pour servir a lhistoire du tapir; et description dune espece nouvelle appartenant aux hautes regions de la Cordillere des Andes. Annales des Sciences Naturelle Zoologie Paris 17:26. Ruez, D. R. 2001. Early Irvingtonian (latest Pliocene) rodents from Inglis 1C, Citrus County, Florida. Journal of Vertebrate Paleontology 21(1):153. Rustioni, M. 1992. On Pliocene tapirs from France and Italy. Bollettino della Societ Paleontologica Italiana 31(3):269-294. Sanders, A. E. 2002. Additions to the Pleistocene mammal faunas of South Carolina, North Carolina, and Georgia. Transactions of the American Philosophical Society 92(5):1. Schultz, C. B., L. D. Martin, & R. G. Corner. 1975. Middle and late Cenozoic tapirs from Nebraska. Bulletin of the Nebraska State Museum 10:1. Sellards, E. H. 1918. The skull of a Pleistocene tapir including description of a new species Annual Report, Florida Geological Survey 10:57. Shunk, A. J. S. Driese, J. Farlow, R. Hulbert, & M. Whitelaw. 2006. High-resolution stratigraphy and sedimentology of late MiocenePliocene paleolacustrine strata deposited in paleosinkhole settings, Eastern U.S.

PAGE 57

HULBERT JR.: Blancan Tapirus from Florida 121 Abstracts with Programs, Geological Society of America 38(7):85. Simpson, G. G. 1929. Pleistocene mammalian fauna of the Seminole Field, Pinellas County, Florida. Bulletin of the American Museum of Natural History 56:561. Simpson, G. G. 1945. Notes on Pleistocene and Recent tapirs. Bulletin of the American Museum of Natural History 86(2):33. Smith, K. S., & R. L. Cifelli. 2000. A synopsis of the Pleistocene vertebrates of Oklahoma. Oklahoma Geological Survey Bulletin, No. 147, 36 pp. Tedford, R. T., X.-M. Wang, & B. E. Taylor. 2009. Phylogenetic systematics of the North American fossil Caninae (Carnivora: Canidae). Bulletin of the American Museum of Natural History 325:1. Tong, H. W. 2005. Dental characters of Quaternary Geobios 38(1):139. Tong, H. W., J. Y. Liu, & L. G. Han. 2002. On fossil remains of early Pleistocene tapir (Perissodactyla, Mammalia) from Fanchang, Anhui. Chinese Science Bulletin 47(7):586 590. Webb, S. D. 1974. Chronology of Florida Pleistocene mammals. Pp. 5 in S. D. Webb, ed. Pleistocene Mammals of Florida. University of Florida Press, Gainesville. Webb, S. D., R. C. Hulbert Jr., G. S. Morgan, & H. F. Evans. 2008. Terrestrial mammals of the Palmetto Fauna (early Pliocene, latest Hemphillian) from the Central Florida Phosphate District. Pp. 293 in X.-M. Wang & L. G. Barnes, eds. Geology and Vertebrate Paleontology of Western and Southern North America, Contributions in Honor of David P. Whistler. Natural History Museum of Los Angeles County Science Series, Number 41. Webb, S. D., & K. T. Wilkins. 1984. Historical biogeography of Florida Pleistocene mammals. Pp. 370-383 in H. H. Genoways & M. R. Dawson, eds. Contributions in Quaternary Vertebrate Paleontology: a Volume in Memorial to John E. Guilday. Carnegie Museum of Natural History, Pittsburgh, Special Publication 8. Witmer, L. M., S. D. Sampson, & N. Solounias. 1999. The proboscis of tapirs (Mammalia: Perissodactyla): a case study in novel narial anatomy. Journal of Zoology 249:249. Wu, X. Z., W. Liu, X. Gao, & G. M. Yin. 2006. Huanglong Cave, a new late Pleistocene hominid site in Hubei Province, China. Chinese Science Bulletin 51(20):2493. A PPENDIX 1 Information on Florida Blancan fossil localities that have produced Tapirus Only approximate location data provided for sites that are still being collected, to protect them from vandalism. Most lack formal stratigraphic names for their sediments, as they are either isolated sinkhole or alluvial deposits. SITES WITH TAPIRUS LUNDELIUSI SP. NOV. 1. Haile 7C (type locality). UF locality AL109, 6 km northeast of Newberry, Alachua County, Florida (Morgan & Hulbert 1995:68; Emslie 1998). Newberry Quadrangle; 29.69 N, 82.56 W. Middle late Blancan, ca. 1.9.2 Ma. 2. Haile 7G. UF locality AL125, 6 km northeast of Newberry, Alachua County, Florida. SE of Quadrangle; 29.69 N, 82.56 W. Middle late Blancan, ca. 1.9.2 Ma. Located about 75 m east of Haile 7G. The lacustrine sediments of the two localities are possibly continuous, but they are currently separated by a covered area. 3. Withlacoochee River 1A. UF locality MR060, in channel of Withlacoochee River about 16 km southeast of Dunnellon, Marion County, Florida. Stokes Ferry Quadrangle; 29.0 N, 82.3 W. Middle late Blancan, ca. 1.9.2 Ma. In situ specimens are the Eocene limestone bedrock. 4. Waccasassa River 9A. UF locality LV040, in

PAGE 58

122 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) bank of Waccasassa River about 8 km southeast of town of Otter Creek, Levy County, Florida. SW of SW Quadrangle; 29.3 N, 82.7 W. Caloosahatchee Formation (or lateral equivalent), middle late to latest Blancan, ca. 1.6.2 Ma. Vertebrate fossils collected interbedded with species of mollusks found in the Caloosahatchee Formation in southern Florida. 5. Inglis 1A. UF locality CI001, 2.5 km southwest of Inglis, Citrus County, Florida. SW of NE Yankeetown Quadrangle; 29.0 N, 82.68 W. Latest Blancan, ca. 1.6.9 Ma. 6. Inglis 1B. UF locality CI013, 2.5 km southwest of Inglis, Citrus County, Florida. SW of NE Yankeetown Quadrangle; 29.0 N, 82.68 W. Latest Blancan, ca. 1.6.9 Ma. 7. Inglis 1D. UF locality CI020, 2 km southsouthwest of Inglis, Citrus County, Florida. NW of SE of section 10, T. 17 S., R. 16 E. on USGS Latest Blancan, ca. 1.6.9 Ma. 8-10. Santa Fe River 8, 8A, and 8C. UF localities CO006B, CO007B, and CO009B, Columbia County, Florida. NW of SE of section 18, T. 7 29.88 N, 82.67 W. Middle late to latest Blancan, ca. 1.6.2 Ma. 11. Walden Pit 2. UF locality SA050, 8 km eastnortheast of Fruitville, Sarasota County, Florida. NE of section 23, T. 36 S., R. 19 E. on USGS Pinecrest Formation; early late Blancan, ca. 2.3 2.6 Ma. SITES WITH TAPIRUS HAYSII 12. Santa Fe River 1 and 1B. UF localities CO003B and CO038B, Columbia County, Florida. NE of NW of section 34, T. 7 S., R. 16 E. on N, 82.70 W. Early late Blancan, ca. 2.3.6 Ma. (UF 177841, 177842, 213920, and 224631). 14. Kissimmee River 2. UF locality OB002, 18 km west of Okeechobee, Okeechobee County, Florida. Section 13, T. 37 S., R. 33 E. on USGS W. Pinecrest Formation; early late Blancan, ca. 2.3.6 Ma. (UF 51250). 15. Kissimmee River 6. UF locality OB006, 18 km west of Okeechobee, Okeechobee County, Florida. Basinger Quadrangle; 27.33 N, 81.03 W. Pinecrest Formation; early late Blancan, ca. 2.3.6 Ma. (UF 52601). This stretch of the river also produced UF 51249, a p2 of Tapirus sp. indeterminate. 16. Inglis 1C. UF locality CI019, 2 km southsouthwest of Inglis, Citrus County, Florida. NW of SE of section 10, T. 17 S., R. 16 E. on USGS Latest Blancan, ca. 1.6.9 Ma. (UF 177844). 17. De Soto Shell Pit 5. UF locality DE018, 17 km south of Arcadia, De Soto County, Florida. NW of SE of section 33, T. 39 S., R. 25 E. on USGS Caloosahatchee Formation, latest Blancan, ca. 1.6 1.9 Ma. (UF 223920, 232075). Neither specimen was collected in situ, and these two specimens could be Irvingtonian rather than Blancan, as this mine also contained sediments from the early Irvingtonian Bermont Formation. 18. Devils Elbow 2. UF locality PU005B, submerged locality on St. Johns River near East Palatka, Putnam County, Florida. Section 38, T. 10 29.65 N, 81.61 W. Nashua Formation, early late Blancan, ca. 2.3.6 Ma. (UF 177843). SITES WITH TAPIRUS SPECIES INDETERMINATE 19. Sommers Pit. UF locality SA029, 8 km eastnortheast of Fruitville, Sarasota County, Florida. NE of section 23, T. 36 S., R. 19 E. on USGS Pinecrest Formation; early late Blancan, ca. 2.3 2.6 Ma. (UF/TRO 1469). 20. Walden Pit 1. UF locality SA049, 8 km eastnortheast of Fruitville, Sarasota County, Florida. NE of section 23, T. 36 S., R. 19 E. on USGS

PAGE 59

HULBERT JR.: Blancan Tapirus from Florida 123 Pinecrest Formation; early late Blancan, ca. 2.3 2.6 Ma. (UF/TRO 1489). 21. De Soto Shell Pit 2. UF locality DE009, 16 km south of Arcadia, De Soto County, Florida. NW of SW of section 27, T. 39 S., R. 25 E. on USGS Caloosahatchee Formation, latest Blancan, ca. 1.6 1.9 Ma. (UF 177836). 22. Lehigh Acres Pit. UF locality LE003, 5.2 km south of Alva, Lee County, Florida. NE of section 26.67 N, 81.60 W. Caloosahatchee Formation, late Blancan, ca. 2.6 Ma. (UF 162666). 23. Haile 15A. UF locality AL032, 4.7 km northeast of Newberry, Alachua County, Florida. NW of Newberry Quadrangle; 29.67 N, 82.57 W. Early late Blancan, ca. 2.3.6 Ma. (UF 17468). 24. US 19 Bridge Site. UF locality GI008, north of bridge over Suwannee River, Gilchrist County, Florida. SW of SW of section 20, T. 10 S., R. 29.6 N, 82.9 W. Early late Blancan, ca. 2.3 2.6 Ma. (UF 247180, 248788, 248789, 255357, 255358). This previously unpublished local fauna also includes Trachemys platymarginata Pseudemys sp., Sternotherus sp., Macroclemys sp., Apalone ferox Hesperotestudo sp., Alligator sp., Dasypus bellus the small morph of Holmesina Paramylodon garbanii Nannippus peninsulatus Odocoileus virginianus and Capromeryx arizonensis 25. Hubbard Pit. UF locality OR003, abandoned clay and sand quarry, Orange County, Florida. SW of NW of section 6, T. 22 S., R. 29 E. N, 81.5 W. Late early or early late Blancan, ca. 2.9.5 Ma. (UF 249089). This quarry produced two fossiliferous horizons with abundant marine invertebrates and vertebrates. The younger bed, which also produced a few terrestrial vertebrate specimens, is of Blancan age based on the presence of Nannippus peninsulatus and molluscan species characteristic of bed 11 from Macasphalt Shell Pit near Sarasota (R. Portell, pers. comm.). In addition to N. peninsulatus and Tapirus fossils of Equus sp. and Dasypus bellus were also recovered from this locality. APPENDIX 2 List of characters and character states used in the phylogenetic analyses. They are adapted from a larger character set originally developed by Hulbert and Wallace (2005), with phylogenetically uninformative characters for the group of species under study here deleted. Original numbering scheme of characters is retained to maintain consistency between publications. 3. temporal crests meet to form sagittal crest: (0) at or posterior to frontal-parietal suture; (1) anterior to frontal-parietal suture. 4. dorsal table of frontal. (0) relatively narrow or small; (1) relatively broad. on same plane; (1) nasals notably stepped down from frontals. 7. size of interparietal bone: (0) large; (1) small. 8. shape of interparietal bone: (0) typically polygonal (hexagonal or diamond shaped); (1) typically triangular. 9. interparietal fusion with occipital: (0) occurs early in ontogeny (before loss of DP4); (1) occurs late in ontogeny (after eruption of P4). 10. nasal length: (0) long, longer than width of combined nasals; (1) short, shorter than width of combined nasals. 14. fossa for meatal diverticulum on nasal: (0) shallow and without distinct margins; (1) deep and with distinct margins. 15. fossa for meatal diverticulum on posterior dorsal surface of nasal: (0) not extensive, does not near midline; (1) very extensive, approaches midline. 16. development of fossa for meatal diverticulum on dorsal table of frontal: (0) very limited; (1) broad exposure with distinct posterior margin. 18. posterodorsal process of maxilla widely

PAGE 60

124 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) exposed dorsal above the orbit, forming base of trough for meatal diverticulum: (0) yes; (1) no. 20. orientation of lambdoidal crests in adults: (0) 22. location of infraorbital foramen relative to the cheekteeth: (0) dorsal to P4; (1) dorsal to P3 or P2. 25. anteromedial process of maxilla: (0) well exposed in lateral view dorsal to premaxilla; (1) covered by premaxilla, not visible in lateral view or barely so. (2) extensive. 27. length of posterior process of premaxilla: (0) long, terminates posterior to P1; (1) terminates dorsal to or just in front of P1; (2) ends about midway over C-P1 diastema; (3) very short, terminates well anterior to midway point of diastema. 28. width of maxillary bar between infraorbital foramen and lacrimal: (0) narrow, usually less than 5 mm; (1) wide, more than 5 mm. 29. shape of lacrimal: (0) narrow, much taller than long; (1) broad, about as long as it is high. concave. 31. anterior lacrimal process(es): (0) absent or very weak; (1) present, well developed. (1) slender, pointed or knobby. 34. typical number of large lacrimal foramen: (0) two; (1) one. 37. relative diastema length: (0) short (ldl/p2m3L 38. location of mental foramen: (0) anterior to the p2; (1) directly ventral to the p2. 39. orientation of anterior margin of ascending ramus of mandible in lateral view: (0) projects vertically and posteriorly, not anteriorly (does not overlie the m3); (1) projects anteriorly as well as vertically below the coronoid process, often lies dorsal to m3. 40. depth of horizontal ramus below m3: (0) shallow (m3depth < 18% mandL); (1) moderate (m3depth between 18 and 19% of mandL); (2) deep (m3depth between 19 and 20% of mandL); 41. relative crown height of cheek teeth: (0) short, relatively brachydont; (1) taller. 43. P1 with: (0) single, small posterolingual cusp (=hypocone of some) and lingual cingulum, but no cross lophs or other cusps; (1) large posterolingual cusp, sometimes with accessory cusps and often with some development of a transverse loph; (2) large posterolingual cusp with strong, complete transverse loph; (3) no distinct posteriorlingual cusp, only a cingulum. 45. P2 AW/PW ratio: (0) on average, less than or equal to 0.85; (1) on average, greater than 0.85. 47. transverse lophs on P2-P4 and p2-p4: (0) poorly separated; 1; well separated. 48. P2 protoloph: (0) does not reach ectoloph; (1) just reaches to base of ectoloph; (2) merges midway or higher onto ectoloph. 49. P3 protoloph: (0) does not reach ectoloph; (1) just reaches to base of ectoloph; (2) merges midway or higher onto ectoloph. 51. P2 metaloph: (0) does not reach ectoloph; (1) just reaches to base of ectoloph; (2) merges midway or higher onto ectoloph. 55. parastyle development on P3-M3: (0) moderate; (1) strong; (2) very strong. 56. labial cingulum on posterior half of upper cheek teeth: (0) present on half or more of P3-M3; (1) absent or very rare on P3-M3. 67. i1 morphology and size: (0) i1 slightly larger than i2, not procumbent; (1) i1 larger than i2, slightly procumbent; (2) i1 much larger than i2, very procumbent and spatulate. 69. length of p2 relative to p3: (0) short, (p2 L)/(p3 L) less than 1.1; (1) long, this ratio greater than 1.1. 72. relative height of unworn protolophid and hypolophid (or hypoconid/entoconid if no hypolophid) on premolars: (0) protolophid

PAGE 61

HULBERT JR.: Blancan Tapirus from Florida 125 distinctly taller than hypolophid; (1) the two are approximately equal in height. 73. cristid obliqua on p3: (0) strong, blocks interlophid valley labially; (1) very weak or absent.

PAGE 62

126 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 49(3) APPENDIX 3. Character state matrix used in phylogenetic analyses. Character numbers in top row correspond to those in Appendix 2. A ? indicates a unknown or non applicable character state. 3 4 6 7 8 9 10 14 15 16 18 20 22 25 26 27 28 29 30 31 Outgroup 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 T. hezhengensis ? ? ? ? ? ? ? ? ? ? ? ? 0 0 1 ? ? ? ? ? T. johnsoni 0 0 0 0 0 ? 0 0 0 0 0 1 0 1 1 1 1 0 0 0 T. webbi ? ? 0 0 0 ? 0 0 0 0 0 0 0 0 0 1 1 0 0 ? T. polkensis 0 1 0 0 1 1 0 1 1 1 0 1 1 0 1 1 1 0 0 0 T. lundeliusi 1 0 0 0 1 0 0 1 1 0 0 1 1 0 1 1 0 0 1 0 T. haysii 0 1 0 0 1 1 1 1 1 1 0 1 1 0 0 2 0 1 1 0 T. veroensis 0 1 0 0 1 1 1 1 1 1 0 1 1 0 0 1 0 1 1 0 T. bairdii ? 1 1 0 1 0 0 1 1 1 1 1 1 0 1 3 0 0 0 0 T. terrestris 1 0 1 1 ? ? 0 0 0 0 0 0 1 1 0 2 1 0 0 1 T. pinchaque 0 0 0 1 0 0 0 0 0 0 1 0 1 1 0 2 1 0 0 1 33 34 37 38 39 40 41 43 45 47 48 49 51 55 56 67 69 72 73 Outgroup 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 T. hezhengensis ? ? ? ? ? 0 0 0 1 1 1 1 0 0 ? ? ? ? T. johnsoni 1 0 1 0 0 2 0 1 0 0 1 1 1 0 0 1 0 0 0 T. webbi 1 0 1 1 0 0 1 1 1 1 0 1 1 1 0 2 0 1 0 T. polkensis 0 1 0 1 0 1 1 0 1 1 1 2 1 1 0 2 0 1 0 T. lundeliusi 0 1 0 1 1 3 1 1 1 1 1 2 1 2 1 2 1 1 0 T. haysii 0 1 0 1 1 3 1 2 1 1 1 2 2 2 1 2 1 1 1 T. veroensis 0 0 0 1 1 2 1 1 1 1 1 2 2 2 1 2 1 1 1 T. bairdii 0 ? 2 1 0 1 1 0 1 1 1 2 1 1 0 2 0 1 ? T. terrestris 1 0 1 1 0 1 1 1 1 1 1 2 1 1 1 2 0 1 1 T. pinchaque 1 0 1 1 0 0 1 0 1 1 0 1 1 1 0 2 0 1 0

PAGE 63

The BULLETIN OF THE FLORIDA MUSEUM OF NATURAL HISTORY publishes original research conducted by our faculty, staff, students, and research associates. We also accept submissions of appropriate, fully funded manuscripts from external researchers. Priority is given to monograph-length papers describing be exceptions as determined by the Managing Editor. Starting in 2010, the Bulletin is published simultaneously in two formats. Approximately 400 printed copies are distributed to libraries and museums world-wide by the University of Florida Library system. Authors have the option of purchasing additional printed copies at cost for distribution to colleagues and associates. An identical, electronic version is posted in PDF format on the Florida Museum of Natural History web site simultaneously with the availability of the printed version and is available free of charge for reading or downloading. Supplemental materials are available only through the web site. INSTRUCTIONS FOR AUTHORS papers must adhere to the rules published in the appropriate international code of systematic nomenclature. RECENT PUBLICATIONS OF THE FLMNH BULLETIN Thomson, A. W., and L. M. Page. 2010. Taxonomic revision of the Amphilius uranoscopus Chelonoidis Elassoma gilberti Zaspel, J. M., S. J. Weller, & R. T. Carde. 2008. A review of Virbia Holomelina Puma concolor A complete list of available issues and current prices of the Bulletin of the Florida Museum of Natural History