Front Cover
 Table of Contents
 Geography and geology
 Brooksville 2 local fauna
 Systematic paleontology
 Megalagus abaconis
 Palaeogale minuta
 Acheronictis webbi n. gen....
 Acheronictis chapini n. gen. et...
 Enhydrocyon cf. pahinsintewakp...
 Osbornodon wangi n. sp.
 Phlaocyon taylori n. sp.
 Age and correlation
 Literature cited
 Back Cover

Group Title: Bulletin of the Florida Museum of Natural History
Title: Brooksville 2 local fauna (Arikareean, latest Oligocene)
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00095780/00001
 Material Information
Title: Brooksville 2 local fauna (Arikareean, latest Oligocene) Hernando County, Florida
Series Title: Bulletin - Florida Museum of Natural History ; volume 43, number 1
Physical Description: 47 p. : ill. ; 28 cm.
Language: English
Creator: Hayes, F. Glynn, 1968-
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2000
Copyright Date: 2000
Subject: Vertebrates -- Florida -- Brooksville   ( lcsh )
Vertebrates, Fossil -- Florida -- Brooksville   ( lcsh )
Paleontology -- Oligocene   ( lcsh )
Genre: bibliography   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Includes bibliographical references (p. 44-47).
Language: Abstract in English and Spanish.
Statement of Responsibility: F. Glynn Hayes.
 Record Information
Bibliographic ID: UF00095780
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 45207686
issn - 0071-6154 ;

Table of Contents
    Front Cover
        Page i
        Page ii
    Table of Contents
        Page 1
        Page 2
        Page 3
        Page 4
    Geography and geology
        Page 5
    Brooksville 2 local fauna
        Page 6
        Page 7
        Page 8
        Page 9
    Systematic paleontology
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
    Megalagus abaconis
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
    Palaeogale minuta
        Page 23
        Page 24
        Page 25
    Acheronictis webbi n. gen. et sp.
        Page 26
        Page 27
        Page 28
        Page 29
    Acheronictis chapini n. gen. et sp.
        Page 30
    Enhydrocyon cf. pahinsintewakpa
        Page 31
    Osbornodon wangi n. sp.
        Page 32
        Page 33
    Phlaocyon taylori n. sp.
        Page 34
        Page 35
        Page 36
    Age and correlation
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
    Literature cited
        Page 44
        Page 45
        Page 46
        Page 47
        Page 49
    Back Cover
        Page 50
Full Text




F. Glynn Hayes

Vol. 43, No. 1, pp. 1-47


L... : -- i' -. -
~ i f l ~ i ; l: ; ^ * ^ ;



are published at irregular intervals. Volumes contain about 300 pages and are not necessarily
completed in any one calendar year.

MARGARET E.B. JOYNER, Managing Editor

ISSN: 0071-6154


Publication Date: July 12, 2000

Price: $6.50

Send communications concerning purchase or exchange of the
publication and manuscripts queries to:
Managing Editor of the BULLETIN
Florida Museum of Natural History
University of Florida
PO Box 117800
Gainesville, FL 32611-7800, U.S.A.
Phone: 352-392-6724
Fax: 352-846-0287
e-mail: mjoyner @ flmnh.ufl.edu


F. Glynn Hayes'


Vertebrate fossils from fissure fill deposits in the Suwannee Limestone located near the town ofBrooksville, Hemando County,
Florida, represent at least 27 mammalian species, constituting the Brooksville 2 Local Fauna. Taxa from the Brooksville 2 Local Fauna
that include first occurrences in Florida are Agnotocastor, Megalagus, Palaeogale, Enhydrocyon, and an Entoptychine rodent. De-
scribed in detail are two insectivores (Centetodon, Parvericius), the lagomorph (Megalagus abaconis n. sp.), the castorid (Agnotocastor
sp.), and all carnivores (Palaeogale, two new musteloids: Acheronictis webbi n. gen. et sp., Arikarictis chapini n. gen. et sp., and three
canids: Enhydrocyon, Osborodon wangi n. sp., Phlaocyon taylori n. sp.). Biochronologic correlation indicates the age of the fauna is
within the undefined biochronologic interval of the "medial" Arikareean NALMA (latest Oligocene). The Brooksville 2 Local Fauna is
correlative with the Cow House Slough Local Fauna and SB-IA Local Fauna.


La fauna local Brooksville 2 esta constituida por f6siles de vertebrados, representando al menos 27 species de maniferos, que
fueron encontrados en fisuras de rocas sedimetarias de Suwannee cerca del pueblo de Brooksville en el condado de Hemando de la
Florida. La fauna local de Brooksville 2 incluye taxones representados por primera vez para Agnotocastor, Megalagus, Palaeogale,
Enhydrocyon y un roedor Entoptychino. Se described en detalle dos insectivoros (Centetodon, Parvericius), el lagomorfo (Megalagus
abaconis n. sp.), el cast6rido (Agnotocastor sp.), y los carnivores (Palaeogale, dos nuevos musteloideos: Acheronictis webbi n. gen. et
sp., Arikarictis chapini n. gen. et sp., y tres canidos: Enhydrocyon, Osbornodon wangi n. sp. y Phlaocyon taylori n. sp.). La correlaci6n
biocronol6gica indica que la edad de esta fauna estA dentro del intervalo biocronol6gico indefinido del Arikareano "medio" NALMA
(Oligoceno mis tardio). La fauna local Brooksville 2 es correlativa con la fauna local de Cow House Slough y con la fauna local SB- A.


Intro d uctio n .......................................................... ............ ......... ... 2
A know ledgm ents ........................................................................... ............................... 4
G geography and G eology ................................................................ ................................. 4
B rooksville 2 Local Fauna ..................................................... ................................... 6
T a p h o n o m y ...................................................... ............ ............................ .................... 8
System atic Paleontology ........................................................ .................................... 10
C en tetodon ... ........... ..... ............................................................................ . ..... .... 10
P a rveric iu s ........................................ .............. ..... .. ................... ..... ........ 12
A gnotocastor sp. ..................................................................... ........................ 14
M egalagus abaconis n. sp.............................................. ........................... 18
P alaeogale m inula ............................................................... ....................... 23
Acheronictis webbi n. gen. et sp........................................................ ............. 26
Arikarictis chapini n. gen. et sp................... .................... ........... 30
Enhydrocyon cf. pahinsintewakpa ................... ..... ............. 31
O sbornodon wangi n. sp.. ................................................. ........................ 32
Phlaocyon laylori n. sp.......................... ......................................... ................ 34
A ge and C orrelation ............................................................. ...................................... 37
S u m m ary .................................................................................... ................. ............ .... 42
L literature C ited ........................................................................... ............................... .. 44

SDepartment of Geosciences and State Museum of Natural History, University of Nebraska, Lincoln, Nebraska, 68588
Hayes, F G. 2000. The Brooksville 2 Local Fauna (Arikareean, latest Oligocene): Hemando County, Florida. Bull. Florida Mus. Nat. Hist. 43(1):1-47.


The Arikareean North American Land Mammal
"Age" (NALMA) spans the late Oligocene-early Miocene
boundary. In Florida this interval has been poorly repre-
sented until recently. In Simpson's (1930) early sum-
mary of Miocene land mammals from Florida, he de-
scribed the first Arikareean fauna from Florida, Franklin
Phosphate Pit No. 2 (FPP2). Two years later, Simpson
(1932) described the famous Hemingfordian Thomas
Farm locality and reviewed the five faunas described in
1930. He suggested that FPP2, on a "necessarily vague
and personal impression might be rather older" than the
other faunas (Simpson, 1932:15). This proved correct,
because FPP2 is still considered Arikareean in age,
whereas the other four localities are presently assigned
to the Hemingfordian or even the Barstovian Land Mam-
mal "Ages."
The second Arikareean fauna came from
Brooksville, Florida, and was briefly described by Patton
(1967), based on four specimens that were "more ad-
vanced than corresponding Late Oligocene forms, yet
more primitive than those of the Middle Miocene."
During the 1970s two more extensive Arikareean
faunas were reported from Florida: the SB-IA Local
Fauna near Live Oak (Frailey, 1978), and the Buda Lo-
cal Fauna from Alachua County (Frailey, 1979). Both
faunas included significant additions to the record of
Florida diversity, and Frailey was able to identify, for
the first time, Arikareean taxa from the Great Plains in
Florida, e.g., Nanotragulus loomisi. In addition, a new
amphicyonid from the Buda Local Fauna, Daphoenodon
notianastes, was recognized in the Brooksville Fauna of
Patton (1967) and in the FPP2 Local Fauna. The pres-
ence of this carnivore, known only from Arikareean sites
in the midcontinent, strengthens the correlation of these
faunas with each other and the Arikareean NALMA.
After the discovery of Buda and SB-IA, MacFadden
(1980) described a small sample from a roadcut in Marion
County, Florida. This site included an Arikareean
oreodont embedded in marine limestone and a few other
isolated specimens that were found in plastic sediments
just above the oreodont. Together these samples make
up the Martin Anthony Local Fauna, one of the few
Florida assemblages that has some lithostratigraphic
control rather than occurring in a karst feature.
The late 1980s brought the discovery of two more
Arikareean faunas: the White Springs Local Fauna and
the Cow House Slough Local Fauna. The White Springs
Fauna is unique in that it is mainly comprised of several
articulated oreodont skeletons preserved in a nearshore


marine facies. Morgan (1989) summarized the oreodonts
and other isolated taxa discovered with them, assigning
a late Arikareean age by comparison with Great Plains'
species. In the same paper, Morgan also first mentioned
the Arikareean Cow House Slough Local Fauna from
Hillsborough County, Florida. Although Morgan (1993)
later discussed Cow House Slough again in more detail
in his summary of mammalian Neogene biochronology
of Florida, this diverse fauna still remains undescribed.
Another important fauna that illuminates the
Arikareean of the Gulf Coastal Plain is the Toledo Bend
Local Fauna of east Texas (Albright, 1991, 1994, 1996).
This fauna is diverse (42 vertebrates, including 27 mam-
mal species). It contains some of the same taxa as Buda,
FPP2, and Cow House Slough.
In his ongoing study of the Toledo Bend Local
Fauna, Albright (1998) recently discussed the Arikareean
faunas in Texas and Florida in relation to the more de-
veloped midcontinental sequences of Arikareean age. In
this paper, Albright reviewed the eight Arikareean Florida
faunas and mentioned that a new rich fauna of Arikareean
age from Florida was currently under study by this au-
The Brooksville 2 Local Fauna represents the ninth
addition to the Arikareean faunas of Florida. It was un-
covered near the town of Brooksville by amateur fossil
hunters of the Florida Paleontological Society in 1994.
Because of their geographic proximity to the town of
Brooksville, this series of samples is called the Brooks-
ville 2 Local Fauna, while the original material described
by Patton (1967) becomes the Brooksville 1 LF.
The Brooksville 2 Local Fauna is, to date, the most
speciose of the Arikareean faunas found in Florida. Its
27 mammalian species rivals the Toledo Bend LF.
Brooksville 2 LF contains more microvertebrate taxa,
whereas Toledo Bend LF has a predominance of larger
ungulates. The new fossils from Brooksville provide a
basis for broader comparisons between the Gulf Coastal
Plain and the classic samples of Arikareean age from
both the midcontinent and the John Day Basin of Ore-
gon (see Tedford et al., 1987, 1996).
This report on the Brooksville 2 Local Fauna pro-
vides: a synopsis of the total vertebrate fauna; a discus-
sion of the taphonomy of the fauna; a detailed system-
atic review of the insectivores, castorids, lagomorphs,
and carnivores; and a section on the age and correlation
of the Brooksville 2 LF with discussion of correlations
within the Arikareean NALMA. Some other components
of the fauna are to be reported on in further detail by
collaborators in the future.
Identification of mammal taxa was based on jaws
and teeth, which provided reliable characters for species

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

Figure 1. Outline map of Florida showing location
of vertebrate local faunas discussed in text. (1) 1-75,
Alachua Co., late Whitneyan or early Arikareean;
(2) Cowhouse Slough, Hillsborough Co., Arikareean;
(3) White Springs, Hamilton and Columbia Co.,
Arikareean; (4) SB-IA "Live Oak," Suwannee Co.,
Arikareean; (5) Buda, Alachua Co., Arikareean; (6)
Brooksville 1, Hernando Co., Arikareean; (7) Frank-
lin Phosphate Pit No.2, Alachua Co., Arikareean; (8)
Martin-Anthony F, Marion Co., late Arikareean; (9)
Thomas Farm LF, Gilchrist Co., Hemingfordian.

designation. However, like most Florida Arikareean fau-
nas, nearly all the fossils consisted of isolated teeth and
disarticulated postcranials. These were sorted prelimi-
narily according to size and general morphology. Com-
parisons were then made with other Arikareean faunas
(using examples of complete or associated specimens
representing the same or closely related taxa) in the col-
lections of the Florida Museum, the American Museum,
and the University of Nebraska State Museum. Musteloid
material was compared at the University of Nebraska,
where casts of relevant European material were avail-
able, and was carried out in Europe by M. Wolsan (writ-
ten communication, 1998) using original material.
Where I have encountered new species, I have named
them utilizing the most diagnostic isolated tooth as the
holotype. Clearly the diagnosis of such species depends
upon the sample, not only the holotype. I have made as-
sociations of isolated material on the following bases:
(1) composite material is assembled on the basis of de-
tailed comparison with more complete similar taxa from
other faunas; (2) size and morphology of a fossil are
consistent with derivation from only a single species; (3)

the material is unique; and (4) the composite material
displays apparent differences relative to other taxa.
All the specimens comprising the Brooksville 2
Local Fauna are curated in the Collection of Fossil Ver-
tebrates, housed in the Florida Museum of Natural His-
tory, Gainesville, Florida. The dental terminology used
for castorids is after Stirton (1935) and Korth (1996).
Terminology for lagomorphs follows that of Wood (1940)
and Dawson (1958). Other dental terminology employs
standard tribosphenic nomenclature (Van Valen, 1966,
fig. 1). Measurements were taken using metric calipers
or, for smaller specimens, a Gertner measuring micro-
scope was used. All measurements are in millimeters
unless otherwise stated. Measurement parameters and
nomenclature follow standard guidelines in the major
published reviews or monographs for each taxon. In
tables and text, a lower case letter indicates lower teeth
(p1-4), while an upper case letter indicates upper teeth
(P1-4). Generic and species names that appear in quota-
tion marks are taxa of questionable or no longer valid
nomenclature retained for purposes of discussion.
Abbreviations used (others defined in figures, tables,

or text): ap, anterior to posterior measurement, length;
alv, alveolus; ca., circa with reference to dates; F, Fauna;
Fm, Formation; LF, Local Fauna; M, sample mean; m#
or M#, molar, lower case for lower molar; N, sample
number; OR, observed range of variation; p# or P#, pre-
molar; tr, transverse measurement, width; tra, transverse
measurement of anterior trigonid width in lower teeth;
trp, transverse measurement of posterior talonid width
in lower teeth; S, standard deviation; UBI, undefined
biochronologic interval.
Institutional abbreviations used: AMNH, American
Museum of Natural History, New York NY; CMNH,
Carnegie Museum of Natural History, Pittsburgh PA;
F:AM, Frick fossil mammal collection, housed in the
American Museum; FMNH, Field Museum of Natural
History, Chicago IL; FSP, Faculte des Sciences, Poiters,
France; LSUMG V, Louisiana State University Museum
of Geosciences, Vertebrates; PU, Princeton University
collection (now at Yale Peabody Museum of Natural
History, New Haven CT); SDSM, South Dakota School
of Mines and Technology, Rapid City; UF, University of
Florida Collections, Florida Museum of Natural History,
Gainesville; USNM, United States National Museum of
Natural History, Smithsonian Institution, Washington DC.


There is a long list of those to whom I owe gratitude
for help with this project. Foremost, thanks to S.D. Webb, as
my advisor, and for originally suggesting this project. Thanks
to B.J. MacFadden, N. Opdyke, and E. Martin for evalua-
tion of the original thesis. For their thoughts and discussion
of this project, I am grateful to R. Hunt, A. Pratt, J. Knight,
R. Portell, B. Albright, X. Wang, and R. Tedford. R McCarty,
A. Poyer, E. Simons, S. Emslie, B. Shockey, L. Cox, A.
Richardson, and the VP class of '94 deserve recognition for
their help in collecting and processing the fossils. E. Taylor
deserves significant thanks as the amateur collector who
brought this site to the attention of the Florida Museum of
Natural History. My appreciation also goes to the Florida Rock
Industries people who freely allowed access to their quarry.
This paper is based on research done as a master's thesis in
the Department of Geology, University of Florida. Support
for this project was provided by the Department of Geology
at the University of Florida, the Florida Paleontological So-
ciety (Gary Morgan Research Grant), and the American
Museum of Natural History (collection study grant).


LOCATION.-The Brooksville 2 Local Fauna
was collected from an abandoned quarry of the Florida


Rock Industries Limerock Mine located in the SW 1/4,
Section 20, Township 21 S, Range 19 E, of the Brooks-
ville NW quadrangle. This is approximately eight kilo-
meters to the northeast of the town of Brooksville,
Hernando County, Florida. The deserted quarry is lo-
cated 30 meters east of Highway 491 A and north of the
intersection of Highway 491A and U.S. Highway 98.
Crews from the Florida Museum of Natural His-
tory identified and collected five fossil terrestrial verte-
brate-bearing sites, labeled IA through IE (see Fig. 2,
inset). These were either from in situ elastic fillings of
karst features or from adjacent spoil piles. Sites IA and
IE were small pockets of laminated clay and sand sedi-
ments filling solution cavities a meter below the present
surface on the southeast wall of the quarry face. Site 1B
was positioned on the quarry floor in a shallow depres-
sion that was seasonally filled with water. Site 1C was a
spoil pile on the west side of a pond in the northern part
of the quarry floor. Site ID was the smallest of the sites,
located at the north end of the excavation along the walls
of an access road. It is uncertain if ID was in place or
disturbed by construction of the road.
All matrix was recovered from sites 1A, 1B, and
IE and screen-washed through three screen sizes to re-
trieve vertebrate fossils. Localities 1C and ID were not
screen-washed because the sediment was not in place
and fossils were scarce relative to sediment volume. The
weights of matrix screen-washed were: IA, 23.78 kg;
IB, 410.31 kg; and IE, 63.71 kg.
REGIONAL GEOLOGY.-The Brooksville 2
LF occurs in the Brooksville Ridge region, the largest of
Florida's central or midpeninsular ridges formed during
Pleistocene sea level fluctuations (White, 1970). The
Brooksville Ridge stretches north-south about 110 miles,
but is divided at the Dunnelon Gap by the westward flow
of the Withlacoochee River. The southern end is marked
by the Zephryhills Gap and the Hillsborough River, and
the northern end is defined by the High Springs Gap and
the Santa Fe River. The Brooksville Ridge has the most
irregular surface of Florida's central ridges, with eleva-
tions ranging from 23 to 66 meters over small distances.
This is due to differential exposure and weathering of
soluble carbonates and resistant siliciclastics. A struc-
tural subsurface high, affecting early Oligocene and older
sediments, is coincident with the Brooksville Ridge. The
high is either the result of a gentle tectonic flexure (the
Ocala Arch, Vernon, 1951) or it is an erosional remnant
of the Ocala and older formations (the Ocala Platform,
Scott, 1992).
The abandoned quarry from which the Brooksville
2 LF was collected consists primarily of early Oligocene
Suwannee Limestone (Yon and Hendry, 1972). The

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

I;c iD 49 1A
0 4 01

Ins et
Brooksvile 2 LF
Location in Quarnyof 1A-1 E

Figure 2. Location map of the Br sve 2 Loca Fauna.

Figure 2. Location map of the Brooksville 2 Local Fauna.

Suwannee Limestone is a marine bioclastic calcarenite
deposited in a shallow shelf environment. Parts of it are
recrystallized and silicified. It has a tendency to become
more arenaceous in its upper sections; the greater influx
ofsiliciclastics reflects renewed uplift in the Appalachians
during the Oligocene (Randazzo, 1997). The recrystal-
lized sections (chert zones) are found in the subsurface
highs of the Suwannee Limestone along the Brooksville
Ridge. The top of the Suwannee Limestone forms an
irregular surface in Hernando County resulting from
uneven karst erosion during the low sea level stand of
the Late Oligocene and Early Miocene (Yon and Hendry,
1972; Haq et al., 1987). This erosional interval may have
formed the sinkholes that were then filled with the sedi-
ments that produced the Brooksville 2 LF.
To further constrain the age of the Brooksville 2
LF, limestone (4 samples) and invertebrate fossils (4
samples) were collected at or near the top of the Suwannee
Limestone exposed in the Brooksville 2 LF quarry. These
samples were then analyzed for strontium isotopic com-
position (87Sr/86Sr) in the Department of Geology at
the University of Florida using standard techniques
(Hodell et al., 1991). The measured isotopic ratios were
converted to age estimates using the regression equa-
tions of Mead and Hodell (1995). The results gave an
age of 35.5-33.3 Ma. This agrees with dates obtained
for Suwannee Limestone elsewhere in Florida by Jones
et al. (1993).

The sediments lying unconformably over the
Suwannee Limestone are unfossiliferous clays and thin
limestones. These have been assigned to the Tampa Fm
or labeled as undifferentiated post-Oligocene (Yon and
Hendry, 1972). Hunter (1970:20) suggested that there is
no "tangible evidence of their proper stratigraphic posi-
tion" even questioning whether they are necessarily of
post-Oligocene age. Following Hunter, I propose to use
the informal name "post-Suwannee" until such time as a
reliable age and position can be established. Because of
the uncertain placement of the post-Suwannee interval,
these sediments cannot be used to help constrain the age
of the Brooksville 2 LF.
The bottom of the quarry exposes the upper part of
the Eocene Ocala Group (Crystal River Fm). The Eocene
Ocala Group is composed of white to gray foraminiferal
and molluscan packstones and grainstones deposited in
a shallow to middle shelf environment (Randazzo, 1997).
The Suwannee Limestone is separated from the Ocala
by an unconformity of uncertain duration. Determina-
tion of the boundary between the Ocala and the Suwannee
is difficult, as the limestone exposed in the quarry is rather
homogeneous in overall lithology. In a section described
by Hunter (1970) from the "Lansing quarry," now the
operational Florida Rock Industries quarry, the
Bumpnose Formation (included in the Crystal River
Formation by Yon and Hendry, 1972) is listed as occur-
ring several feet below the chert boulder zone of the


Suwannee Limestone. This chert zone is present in the
abandoned quarry two or three meters above the quarry
floor, and that is the lithological evidence that the Crys-
tal River Formation is present.
More generally, in central Florida the three main
lithological units described above are variable in their
occurrence and superpositional relationships. In many
areas the Suwannee Limestone is missing, either through
erosion or nondeposition, and the Ocala Group directly
underlies the post-Suwannee deposits. Along the crest
of the Brooksville Ridge, and in many parts of central
Florida, the Suwannee and post-Suwannee units are both
missing, leaving only the Ocala Group at the surface.


Table 1 is a preliminary list of the taxa comprising
the Brooksville 2 Local Fauna. I am grateful to the col-
leagues acknowledged above for allowing me to cite their
preliminary results, including the rodents, bats, reptiles,
and amphibians. Presently, 46 species of vertebrates are
listed in the fauna. Fishes are represented by two species
of sharks (Chondrichthyes) and one of teleost
(Osteichthyes). These are typical of the Oligocene ma-
rine Suwannee Limestone and are a reworked compo-
nent of the fauna. Reptilia and Amphibia are very abun-
dant in the fauna with at least three species of frogs
(Anura), two salamanders (Urodela) including a siren, a
small crocodile, a mud turtle (Kinosteridae), boid snakes
(Serpentes), and three species of lizards (Squamata).
Birds are rare in the fauna with only three postcranial
bones representing the Galliformes. The mammalian taxa
include a mouse-sized marsupial, two insectivores, three
families of bats (Chiroptera), numerous rodents of seven
families (Geomyidae, Sciuridae, Heteromyidae,
Heliscomyidae, Florentiamyidae, Castoridae, Eomyidae),
a rabbit (Lagomorpha), seven species of carnivores
(Musteloidea, Canidae), one horse (Equidae), and three
kinds of artiodactyls (Camelidae, Hypertragulidae,
Merycoidodontidae). Figure 3 shows the relative abun-
dances of mammal orders in the fauna, based on the mini-
mum number of individuals for each taxon.
Over 60 teeth and a few jaws belong to the marsu-
pial species. This sample is unusually large, as marsu-
pials are generally rare in most localities (Korth, 1992).
Analysis of the marsupial (Hayes and Wolff, in prepara-
tion) indicates that this material is a new species of
Herpetotherium, possibly a descendant of "Peratherium"
listed in the Whitneyan 1-75 LF (Patton, 1969). Both these
marsupials appear to be closely related to H. merriami
(Stock and Furlong, 1922) from the John Day fauna of


TABLE 1.-Preliminary Brooksville 2 Local Fauna
vertebrate faunal list.
Ginglymostoma sp.
Carcharhinus sp.
Teleostei: undet.
Anura: 3 Families present




Sirenidae: n. gen. et sp.
Notothalmus sp.


Calamagras : 2-3 species, 1 n. sp.
Ogmophis sp.
Geringophis n. sp.

Galliformes: indet.
Herpetotherium n. sp.
Parvericius montanus*
Centetodon magnus*
Emballonuridae: n. sp.
Mormoopidae: 2 n. sp.
Vespertilionidae: n. sp.
Heteromyidae: 2 sp.
Entoptychinae: n. gen. et sp.
Agnotocastor sp. *
Megalagus abaconis n. sp.*
incertae sedis
Palaeogale minutus*
Acheronictis webbi n. gen. et sp.*
Arikarictis chapini n. gen. et sp.*
Enhydrocyon cf. pahinsintewakpa*
Osbornodon wangi n. sp*
Phlaocyon taylori n. sp.*
Miohippus sp.
Nothokemas waldropi.
Nanotragulus loomisi
Phenacocoelinae: genus indet.
* Mammalian taxa described in detail in this paper.

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hemando County, Florida

Figure 3. Relative abundances of mammal orders from the Brooksville 2 LF. Based on minimum number of individuals.

Oregon rather than to the Great Plains lineages of
Herpetotherium (Green and Martin, 1976; Korth, 1994).
Two distinct genera, the immigrant hedgehogs
Parvericius and Centetodon, represent insectivores.
Centetodon is a common insectivore of the early Oli-
gocene, but much rarer in later faunas (Lillegraven et
al., 1981). Its terminal species is found in the Brooksville
2 LF and in the Buda LF.
The bats are some of the best-preserved material in
the fauna with three families and four species present.
Morgan's (pers. comm., 1996) preliminary work indi-
cates that these are all undescribed species. The most
common bat occurring in the fauna belongs to the
Mormoopidae; the family at present has its northern
limit in southernmost Arizona and Texas and ranges
through Central America into southern Brazil (Nowak,
1999). The other two families are the Vespertilionidae
and the Emballonuridae. Vespertilionids are found
worldwide and have the widest ecological range of
any bats, while the emballonurids, like the
mormoopids, are restricted to the tropics and subtrop-
ics (Nowak, 1999).

Rodents are the most abundant vertebrates of the
Brooksville 2 LF. Over seven families are present, with
more than one species each in the Heteromyidae,
Eomyidae, and Sciuridae. The geomyid belongs to the
subfamily Entoptychinae that includes the genera
Pleurolicus, Gregorymys, Entoptychus, and
Schizodontomys. Gregorymys and Entoptychus are im-
portant as biostratigraphic indicators in the John Day
beds in Oregon and the Great Plains (Rensberger, 1971,
1973, 1983; Tedford et al., 1987). The Brooksville
entoptychine is closest to Gregorymys in morphology in
that it is relatively low crowned with complex premolars,
but it has several dental features in common with
Pleurolicus and Entoptychus as well. Therefore, assign-
ment to any known genus is doubtful. Castorids are rep-
resented in the fauna by Agnotocastor sp., which is fully
described in the systematic section.
The carnivores are diverse and consist of a wide
range of body sizes and ecotypes. The first definitive
Arikareean Gulf Coast sample ofPalaeogale occurs at
Brooksville 2 with some 13 specimens, including sev-
eral dentaries. Palaeogale is a small mustelid-like taxon,


Lagomorpha :.:.:

but it is uncertain if it is more closely related to the
feliforms than to the caniforms. This species has a
hypercarnivorous dentition and presumably had a diet
similar to that of modern weasels consisting almost en-
tirely of small mammals.
There are two new genera and species ofmusteloids
with mesocamivorous dentitions suggesting that they
were omnivorous and, like many present-day mustelids,
fed on a variety of foods depending on the season (Ewer,
1973). The first new musteloid that is present is the same
size as Palaeogale and may be similar to ?Plesictis sp.
described by Macdonald (1970) in the Wounded Knee
Fauna. Four teeth represent a second new, mink-sized
genus similar to the European genera Angustictus and
The most abundant carnivore is Phlaocyon.
Phlaocyon is a small raccoon-sized borophagine canid
with a hypocarnivorous dentition. A second large carni-
vore at Brooksville 2 is the coyote-sized hesperocyonine
canid Osbornodon, which has a hypocarnivorous denti-
tion with an enlarged M2. The largest carnivore in the
fauna is referred to Enhydrocyon although its presence
is recorded from a single tooth. A large hypercamivorous
carnivore is not present in this fauna.
The equid found in the Brooksville 2 LF is the same
species as occurs at Cow House Slough LF of Florida.
Morgan (1993) referred the Cow House Slough LF equid
toMiohippus sp.; however, this is questionable because
the ectoloph connects to the metaloph on the molars, a
derived trait not often found in Miohippus. Prothero and
Shubin (1989) listed this character as a synapomorphy
uniting the Anchitheriinae with the Equinae, subfamilies
of later more derived horses. The most advanced de-
scribed Miohippus, M equinanus, only occasionally
shows this condition. The Cow House Slough and
Brooksville 2 equids appear at the time of transition be-
tween earlier Miohippus and later Archaeohippus and
Parahippus (horses common in the Thomas Farm LF)
(Prothero and Shubin, 1989). The Brooksville 2 and Cow
House equids probably are transitional forms that have
features of both advanced Miohippus and primitive
Archaeohippus. However, because the Brooksville 2 and
Cow House Slough teeth show more primitive charac-
ters, like absence of plications on the metaloph and more
conical protocones, these teeth are conservatively referred
to Miohippus.
The three artiodactyls preserved in the fauna are a
rare oreodont, a small species of Nanotragulus, and a
small common species of Nothokemas. The Brooksville
2 oreodont is close to the oreodont in the White Springs
LF (Morgan, 1989). The taxonomy oforeodonts is largely
determined by cranial characters, and with only sparse


dental material from Brooksville 2 does not warrant an
assignment. The Nanotragulus material falls, like the
Buda Nanotragulus, within the smallest size group (IV)
of Frick (1937). Included in this group are the species N
loomisi, N. lulli, and N. inlermedius. Frailey (1979) ar-
gued that these species were synonymous and by prior-
ity all should be placed in N loomisi. Tedford et al. (1996)
recognize Nanotragulus loomisi as first appearing at the
beginning of the Arikareean and extending into "medial"
Arikareean. The Buda dentitions differ from the
Nanotragulus from Brooksville 2 in having more elon-
gate teeth but similar widths. The Brooksville 2
Nanotragulus may be a more primitive population, but
following Frailey (1979) I have referred it to N. loomisi.
The third artiodactyl found at Brooksville 2 repre-
sents Nothokemas waldropi. In his study of SB-1A,
Frailey (1978) named Nothokemas waldropi as the small-
est species of the genus and suggested that this new spe-
cies was an Arikareean ancestor to the Hemingfordian
Floridatragulus and Nothokemas floridanus. Later, in
his paper on the Buda LF, Frailey (1979) referred small
artiodactyl material in the fauna to Camelidae n. gen. et
sp. He refrained from further analysis of the sparse
sample; however, he did suggest that this sample was
allied to a group comprised ofNothokemas, Oxydactylus,
Gentilicamelus, Floridatragulus, and Miolabis. Com-
parison of Nothokemas waldropi with the Buda
Camelidae n. gen. et sp. and the Brooksville 2
Nothokemas indicates that they all are members of the
same genus, if not species. There is a small size differ-
ence among samples from the faunas, yet their overall
morphology is similar, hence the referral of the
Brooksville 2 camel to N waldropi. In addition to the
above local faunas, the small species of Nothokemas is
also found in the Cow House Slough LF and the Franklin
Phosphate Pit No. 2 LF. With these new occurrences,
Nothokemas is recognized as the most common artio-
dactyl of the Florida Arikareean.


The Brooksville 2 Local Fauna was partially re-
covered from a series of karst deposits in the Suwannee
Limestone. Animals can use such features as living quar-
ters or karst may form deadly traps (Behrensmeyer and
Hook, 1992). This taphonomic analysis relied on five
lines of evidence: physical condition of the bones, rela-
tive representations of the bones, faunal composition,
bone orientation, and sediment composition (Pratt, 1989).
The majority of the postcranials recovered show sev-
eral types of physical modification. Many of the larger
bones exhibit evidence of processing by carnivores with

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

punctures and other direct signs of mastication. Some of
the rodent teeth and small carnivore teeth show evidence
of acid etching in digestive tracts. The enamel of these
teeth has a powdery appearance, whereas teeth abraded
by water tend to have a polished surface (Pratt, 1989).
In all the sites bones show some evidence offluvial trans-
port. Site IB in particular has bones that exhibit exten-
sive modification by water abrasion, scoring a stage 3 of
Hunt's (1978) classification, where 0 is the least worn
and 3 is the most worn. Sites IA and IE show the least
amount of transport with bones classified as less than 1.
None of the postcranial elements were articulated and
much of the dental material consisted of single teeth with
the roots broken. Teeth from site IB were significantly
more degraded than those in IA and IE. Sites IA and
IE had the best-preserved materials, which include com-
plete mandibles retaining teeth, and many complete
postcranials of the smaller vertebrates. The majority of
the complete jaws and also the intact postcranials be-
long to the chiropterans. This suggests that they were
transported very little compared to the other mammals
in this deposit. In contrast, most of the larger mammals
were represented by extensively broken skeletal material
and isolated teeth.
The relative representations of skeletal elements lead
to some general conclusions. Long bones, such as tibiae
and humeri, were present only as distal or proximal ends.
All the complete bones of larger mammals consisted of
the most durable compact postcranials such as astragali,
carpals, and phalanges. Cranial material, which is often
the most delicate of skeletal material, was rare.
Faunal composition shows several different patterns
in age distribution, size distribution, and taxa present.
The rodents are represented by all age groups. Rodent
teeth ranging from very slightly worn deciduous pre-
molars, through unworn molars, to molars that had worn
down through all the enamel have been recovered. The
bats tend to be younger animals with lightly worn teeth.
The rabbits, beavers, and larger mammals display a broad
distribution of age classes, but the majority are either
young or old animals indicated by unfused epiphyses of
the postcranials, deciduous premolars, and unerupted
molars for the young individuals, or by heavily worn
teeth and degraded skeletal articular surfaces for the old.
The faunal remains at Brooksville 2 show an ab-
sence of such larger herbivores as rhinos, anthracotheres,
chalicotheres, and tapirs that are often found in similar
age faunas of the Great Plains and in some of the other
Florida Arikareean faunas. Large carnivores are also
absent from Brooksville.
The species present in the fauna also provide infor-
mation as to the mode of occurrence of the bone accu-

mulation at the sites. The chondrichthyans are typical of
those found in the early Eocene and are therefore a re-
worked component. Sirenians and kinosternid turtles are
exclusively aquatic animals, so their presence along with
frogs and crocodilians indicates the existence of nearby
water either as lacustrine sinks or cave ponds. The abun-
dant chiropterans are often common cave dwellers, which
suggests that the karst features were more cave-like than
open-sided sinkhole traps. The near absence of birds
could be taken as further evidence that this fauna accu-
mulated in caves rather than in an open sink. The num-
bers of predator species relative to prey species is high
in the Brooksville Fauna with seven carnivores and ap-
proximately twenty possible prey taxa. This suggests that
at various times caves and hollows in the limestone were
used as carnivor denning sites.
From the above evidence, several conclusions about
the accumulation of the Brooksville 2 LF can be drawn.
The animals of the fauna are parautochthonous, that is
they have been "transported from the death or discard
site but have stayed within the original habitat"
(Behrensmeyer and Hook, 1992:19). Most of the larger
ungulate species, Nanotragulus, Nothokemas,
Miohippus, and some of the smaller taxa are the result
of accumulation by carnivores. Absence of the relatively
very large ungulates is due to the limiting factor of prey
size that confronted the carnivores utilizing the karst
The overall importance of carnivore activity is dif-
ficult to gauge, because some modifications, such as dis-
articulation and preservation of the denser skeletal ele-
ments, can be produced either by water transport or by
carnivores. In view of this ambiguity, the age distribu-
tion of the rodent population, and the lack of digestive
modification of many of the rodent teeth, it is proposed
that many of the smaller taxa of the fauna, including the
bats, actually inhabited various parts of the karst sys-
tem. The evidence of the increased degradation of the
samples from 1B (located in the bottom of the quarry) in
comparison with IA or IE (located near the top of the
quarry) and that bones from all the sites shows evidence
of water transportation implies that the Brooksville 2
sites were catchment basins within a larger system of
anastomosing solution pipes. Site IB has the lowest ele-
vation of these basins. This system of pipes may also
have acted as a taphonomic filter preferentially trans-
porting certain elements, either through size restriction
of the karst pipes or as a result of hydraulic transport
factors. Such mechanisms account for the great richness
of these samples in relation to the volume of surround-
ing sediments.
Site 1B was submerged during most of the collect-

ing trips, so excavation of the site was conducted blindly
and stratigraphic information was not preserved. Screen-
washing provided abundant evidence that the material
from that site underwent the most transportation and
abrasion. Site IB contained the largest and most numer-
ous clasts of the surrounding limestone, and the largest
percentage of reworked fossils of any of the Brooksville
2 sites. These clasts were subrounded to rounded, whereas
1A and 1E contained more angular detritus.
Sites 1A and 1E were collected in place, so their
stratigraphy was preserved. The sediments were depos-
ited in thin laminae, with layers of alternating sand- or
silt-size particles and brown to greenish clays. The fos-
sils were preserved in association with the sandier lenses
rather than heterogeneously mixed within the sediments.
Grain size analysis showed that 91% of the sediment
was greater than medium silt to clay size (6) of the
Udden-Wentworth scale). Medium to fine (3-5() well-
rounded quartz sand comprised 4% of the total weight.
Limestone pebbles to granules (-1 to -3<) comprised
another 1%. Fossil vertebrate remains and fragments of
reworked echinoids (-4 to -5k) comprised the final 4%
of the total weight. This evidence suggests that at times
of higher water flow the vertebrates were washed in along
with the sand and silt derived from the calcarenaceous
Suwannee Limestone, and then subsequently covered by
clay derived from weathering of the limestone. The clay
settled from suspension during times of lower flow or
from standing water.
Site 1C was a spoil pile, so no original features
were left. The sediments that contained the fossils con-
sisted of grayish clay and pebble- to gravel-size chert;
very little limestone was present. Site 1C was the only
site that produced intact associated materials, like the
Osbornodon palate. Clearly, the taphonomic processes
that formed 1C were different from the other sites, but
because the sediment was displaced its origin is dubi-
ous. While sediments from site ID were similar to those
of IA, 1B, and IE, they did not appear to be in place, so
this site also was unsuitable for stratigraphic or sedi-
ment analysis.
In summary, the Brooksville 2 LF animals used a
system of karst features as dens or burrows in which
they and their prey accumulated. As the karst terrain
evolved some of the remains were transported through
solution pipes to be redeposited subaqueously in the larger
fissures of sites 1A, IB, and IE. There is no compelling
evidence to indicate that the fossil accumulation of any
of these sites spanned a significant interval of geologic
time. The facts that the taxa are all similar for each of
the sites and variation within the more numerous taxa is
not broad implies that the time of accumulation was of


relatively "short" duration, less than a million years. In
this regard, Brooksville 2 LF resembles most of Florida's
other early karst sites (Patton, 1969; Frailey, 1978, 1979;
Pratt, 1990).

Order INSECTIVORA Illiger, 1811
Superfamily SORICOIDEA Fischer de Waldheim,
Family GEOLABIDIDAE McKenna, 1960
Genus CENTETODON Marsh, 1872
Centetodon magnus Clark, 1936
Figure 4, Table 2

TYPE.-PU 13835, right mandibular fragment
with p4-m3
OCCURRENCE.-Uintan-late?Arikareean, Great
Plains, Wyoming, and Montana; "medial" Arikareean
UBI, Florida
UF 163561 Lm2
DESCRIPTION.-The M2 from Brooksville 2 LF
is 4-rooted, broad transversely, with strong anterior and
posterior cingula and broad stylar shelf. The metacone
and paracone are subequal in size but the paracone is
taller. The protocone is wide and as tall as the paracone.
The anterior slope of the protocone is steeper than the
posterior slope. The postprotocrista is lower than the
preprotocrista and separated from it by a deep talon ba-
sin. The parastyle is the largest of the stylar cusps, stand-
ing separately from the paracrista that ends at the base
of the parastyle. The parastyle is joined with a broad
paracingulum that extends upward to become the
preprotocrista. A small metastyle is confluent with the
metacrista. A strong metacingulum widens between the
metastyle and the metacone, thins at the metacone, and
joins with the postprotocrista. A metaconule is present
midway along the post protocrista where a small ridge
from the base of the metacone extends to and merges
with the postprotocrista. The paraconule is absent. The
posterior cingulum is the widest of the cingula surround-
ing the protocone, beginning at the lingual base of the
metacone and ending at the hypocone. The anterior cin-
gulum begins at the lingual base of the metacone and
ends in a small cuspule. These cingula do not join around
the lingual tip of the protocone. The two lingual roots
extend upward subparallel to each other. The hypoconal
root is the larger of the two lingual roots, while, of the
two buccal roots, the paraconal root is the larger. The
buccal roots are located on the anterior and posterior
comers of the stylar shelf.

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida


Figure 4. Centetodon magnus, x12, UF 156389, RM2. (A) lingual view; (B) occlusal view.

The only lower tooth of Centetodon from
Brooksville 2 is a heavily worn m2. The trigonid is over
twice as tall as the talonid. The talonid width equals that
of the trigonid but is anteroposteriorly compressed
compared to the trigonid. The metaconid and protoconid
are of equal height, although the protoconid is more
massive. The paraconid is the smallest trigonid cusp. It
lies close to the metaconid and is separated from it by a
slight furrow. An anterior cingulid begins on the ante-
rior face of the trigonid and descends steeply to the base
of the protoconid where it joins with a weak labial
cingulid. The labial cingulid becomes stronger at the
hypoflexid. The labial cingulid ends at the anterior base
of the hypoconid. The talonid cusps are worn to the point
that they form a ridge around the talonid basin. The
entoconid region protrudes posterobuccally. There is no
posterior cingulid. The cristids are parallel and slanted
buccally from the trigonid. The cristid obliquid starts

TABLE 2.-Dental measurements of Centetodon magnus from
Brooksville 2 LF.

UF 156389 RM2 UF 163561 Lm2
ap 1.75 ap 1.67
tra 2.98 apt .87
trp 2.74 tra 1.32
trp 1.05

from the hypoconid region and joins the wall of the tri-
gonid at the middle base of the protoconid. The hypo-
flexid forms a steep sloped notch ending at the labial
DISCUSSION.-The Brooksville 2 specimens
match morphologically and fall within the size range for
Centetodon magnus (Lillegraven et al.,1981). In the
Brooksville 2 species the lingual roots are not widely
spread and are almost parallel. The only two species of
Centetodon with this morphological character are C.
magnus and C. hendryi (Lillegraven et al., 1981, Fig.
10). C. hendryi is known from Eocene deposits and is a
smaller species.
The two Centetodon species that range into the
Arikareean are C. magnus and C. divaricatus (Korth,
1992). C. divaricatus is larger than the Brooksville 2
species and is similar to C. wolffi from the Whitneyan
and Orellan. Korth (1992) described C. divaricatus as
the largest species of the genus based on his comparison
with C. magnus, but failed to include C. wolffi, which is
larger in all parameters than C. divaricatus. Both C.
divaricatus and C. wolfi have widely spread lingual roots
as in C. marginalis but unlike C. magnus and the
Brooksville 2 sample.
Albright (1998) listed Centetodon from both Cow
House Slough and Buda. The Buda specimens, an upper
molar (UF97380) and a complete lower molar (UF97381)
are also referable to C. magnus. The evidence presented
here and in Albright (1998) constitutes the first record
of Centetodon from east of the Mississippi.

Superfamily ERINACEOIDEA Fischer de Waldheim,
Family ERINACEIDAE Fischer de Waldheim, 1817
Subfamily ERINACEINAE Fisher de Waldheim,
Tribe AMPHECHININI Rich, 1981
Genus PARVERICIUS Koemer, 1940
Parvericius montanus Koerner, 1940
Figure 5, Tables 3, 4

TYPE.-YPM 13956, right maxilla fragment with
OCCURRENCE.-Arikareean-late Barstovian,
Great Plains, "medial" Arikareean UBI, Florida
UF163564 RM1, UF163565 Rml, UF163566 Rml,
UF163567 LM2, UF163568 RMI, UF163569 LM1,
UF163570 RM1, UF163571 Lm2, from Brooksville 2
LF, Florida
DESCRIPTION.-The Ml is wider than long. The
metacone and protocone are subequal and the largest
cusps. The paracone and hypocone are of similar size.
The metacone is the tallest cusp. The hypocone is the


lowest. The protocone lies slightly posterior to the para-
cone on the anterior margin. The distance between the
protocone and paracone equals that between the meta-
cone and hypocone. A well-developed postmetacrista is
directed from the cusp of the metacone to a strong
metastylar spur. A slight preparacrista extends from the
paracone to merge with a comparatively weaker
parastylar spur. There is an almost complete marginal
cingulum except for the lingualmost parts of the proto-
cone and hypocone where the cingulum grades into the
stylar spurs. The paracone and protocone are linked by
a preprotocrista that gently slopes from the cusp of the
paracone to the cusp of the protocone. A less well devel-
oped postprotocristajoins the metacone to the protocone
and is divided medially with a posterolingually oriented
crista extending to the hypocone. The postprotocrista ends
at the base of the metacone. No metaconule or paraconule
is visible, possibly due to the wear stage of the tooth.
Three roots are present above the tooth. The lingual root
is the largest, anteroposteriorly flattened and trending to
the margins of the crown. The buccal roots.are equal in
size and are situated almost at the buccal margin of the
paracone and metacone.
On M2 the protocone, metacone, and paracone are
subequal. The hypocone is small and more distinct com-
pared to that on the Ml. There is no ectoflexus in the


Figure 5. Parvericiusmontanus, x12. (A) UF163569, LM1, occlusal view; (B) UF163569, LMI, lingual view; (C) UF 163567,
LM2, occlusal view; (D-F) UF163566, Rml; (D) lingual view; (E) occlusal view; (F) labial view.

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

TABLE 3.-Dental measurements of Parvericius montanus from Brooksville 2 LF

Ml (UF163563)
ap 2.18
tr 2.66



(UF 163569)


N=5 ap M=2.05 S=0.09
tr M=2.44 S=0.18

M2 (UF 163565)
ap 1.60
tr 2.08

(UF 163566)

m2 (UF 163571)


buccal margin and, unlike the M 1, the buccal margin is
slanted mediolingually rather than subparallel to the
maxilla. The metacone is moved posteriolingually closer
to the hypocone. There is no metastylar spur; the
parastylar spur projects buccally rather than anteriorly
on the M and is now dominant. A paraconule may be
present as a thickening in the preprotocrista. A meta-
conule is part of the postprotocrista. There is a small
crista that runs from the base of the hypocone to the
metaconule. There are three roots similar to the M ex-
cept that the lingual root is moved above the protocone
instead of being equally spaced between the protocone
and hypocone, as on the M1.
The trigonid is a little over half the length of the
ml. The metaconid is the tallest cusp, the protoconid is
the largest, and the paraconid the smallest. The proto-
conid is pointed anteriorly against the buccal margin. A
deep cleft, almost to the base of the crown, separates the
metaconid and paraconid from the protoconid. This cleft,
on the buccal side, is closed by a paralophid joining the
metaconid with the protoconid. A large entoconid and a
smaller hypoconid are the only cusps of the talonid. The
postcristid runs lingually from the hypoconid to the base
of the entoconid. A narrow buccal cingulid starts from
the base of the protoconid and continues to the base of
the hypoconid where it joins with a postcingulum. The
postcingulum extends upward lingually from the buccal
cingulid to merge with the entoconid just above the
The m2 is similar to the ml except it is 25% smaller
and more anteroposteriorly compressed than ml. The

paraconid is reduced and is not distinct from the
paralophid. The entoconid and hypoconid are more com-
pressed buccolingually than on the ml making the tal-
onid more quadrate.
DISCUSSION.-These teeth from Brooksville 2
compare in size and morphology with those of
Parvericius montanus published by Rich and Rasmussen
(1973) and Rich (1981). Korth (1992) described a new
species ofParvericius, P voorhiesi, that is the same size
as P montanus. The new species differs in that the m3
has a posterior cingulid and the P4 has a well-developed
parastyle. Since the Brooksville 2 sample contains nei-
ther of these diagnostic features, recognition of this taxon
is not possible. Korth states that other specimens of the
Arikareean age, namely those described by L. Macdonald
(1972), might also represent P voorhiesi, but there is no
way to know without the defining teeth. I have taken a
more conservative approach, questioning the validity of
the new species, and assign the Brooksville 2 sample to
P montanus.
Rich and Patton (1975) described a hedgehog from
the Buda LF as Amphechinus sp. because it compared
well in size and morphology to the m2 of Amphechinus
horcloudi. They showed that it was separable from other
taxa by describing key differences in similar-sized mam-
mals such as bats and shrews as well as other genera of
hedgehogs. However, they failed to make a comparison
with Parvericius. Although some workers consider
Amphechinus and Parvericius to be synonymous
(Macdonald, 1972), Rich recently suggested these taxa
are separate. In comparison with M from Parvericius,


TABLE 4.-Comparison of dental measurements of middle Tertiary Erinaceine hedgehogs with
Brooksville 2 LF sample. Compiled from Korth 1992; Rich and Rasmussen 1973; Rich 1981.

Amphechinus Brooks-
Stenochinus Paleoscaptor horncloudi P voorhiesi P montanus ville 2 LF Buda

M1 ap 1.7 2.2 3.1-3.4 2.00-2.15 2.0-2.2 1.95-2.18
tr 2.2 2.6 3.4-3.7 2.38-2.42 2.1-2.6 2.2-2.66

M2 ap 1.2 1.6 x x 1.5-1.7 1.6
tr 1.9 2.2 x x 1.8-2.2 2.08

ml ap 2.0-2.2 2.7 3.6 1.96-2.30 1.9-2.5 2.52-2.62 2.7
apt 1.1-1.4 1.6 2.1 x 1.0-1.5 1.51 1.6
tra 1.1-1.6 1.5 2.3 1.21-1.46 1.2-1.6 1.48-1.54 1.7
trp 1.1-1.7 1.6 2.3-2.4 1.12-1.49 1.3-1.8 1.72 1.6

m2 ap 1.8 1.9 2.7-3.4 1.67-1.90 1.5-1.9 1.92
apt 0.9-1.1 1.1 1.7-2.0 x 0.8-1.0 1.12
tra 1.3-1.4 1.4 1.9-2. 1.22-1.26 1.0-1.4 1.20
trp 1.2-1.3 1.2 1.6-2.1 1.15 0.9-1.4 1.08

the Buda tooth is only slightly longer and matches in
other measurements (see Table 4). Since the Brooksville
2 species has upper molars that are clearly smaller than
those ofAmphechinus, I assign this erinaceid m2 from
Buda to Parvericius as a m 1.
Rich (1981) separates Amphechinus and Parvericius
becauseAmphechinus is larger and has an anteroposteriorly
elongated trigonid on ml, whereas Parvericius is
anteroposteriorly compressed. While it is clear that there is
a size difference between the two genera (Table 4), com-
parisons of measurements of the ratio of overall length to
length of the trigonid do not support the second diagnosis.
Parvericius has a ml trigonid that is 50-60% of overall
length and Amphechinus, with fewer specimens, has a ml
trigonid that is 58-60% of overall length. These overlaps in
these proportions weakens the distinction between the two
genera. I have not proposed formal synonymy here but rec-
ommend further investigation of the diagnoses separating
Parvericius and Amphechinus.

Order RODENTIA Bowdich, 1821
Superfamily CASTOROIDEA Gill, 1872
Family CASTORIDAE Gray, 1821
Subfamily AGNOTOCASTORINAE Korth & Emry,
Genus AGNOTOCASTOR Stirton, 1935
Agnotocaslor sp.
Figure 6, Table 5

TYPE OF GENUS.-Agnotocastor praete-
readens, Stirton, 1935, AMNH 1428
OCCURRENCE.-Chadronian-early Arikareean,
Great Plains; "medial" Arikareean UBI, Florida
2, UF143320 Rm2, UF143321 Lmlor 2, UF143684
Lmlor 2, UF163601 RP4, UF163602 RP4, UF163603
RP4, UF163604 Rp4, UF163605 RP4, UF163606 Lp4,
UF 163607 Rmlor 2, UF163608 Rmlor 2, UF163609
Rmlor 2, UF163610 Rmlor 2, UF163611 Lmlor 2,
UF163612 Rml, UF163613 Lmlor 2, UF163614 RM2,
UF163615 RMlor 2, UF163616 RMlor 2, UF163617
RMlor 2, UF163618 LMlor 2, UF163619 RM3,
UF163620 Rp4, UF163621 Lp4, UF163622 Ldp4,
UF163623 RM3, UF163624 Ldp4, UF163625 Rdp4,
UF163626 Rdp4, UF163627 LM3, UF163628 RM3,
UF163629 LM3, UF163630 LM3, UF163631 Rdp4,
UF163681 upper incisor, UF163752 Lm3, UF163753
M?, UF163754 RM3, UF163755 RMlor 2, UF163756
Rdp4, UF163757 RM3, UF163758 RMlor 2, UF163759
LMI, UF163760 M?, UF175401 lower incisor,
UF175402 lower incisor, UF175403 upper incisor,
UF175404 upper incisor: from Brooksville 2 LF. UF
40189 upper molar: from Cow House Slough LF.
DESCRIPTION.-This large sample of cheek
teeth shows that they are consistently low-crowned. The
fossettes are irregularly walled and complex in unworn
specimens, becoming smoother and less complex as wear

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

progresses. The premolars are the largest teeth, while
the molars decrease in size from Ml/(ml) to M3/(m3).
The incisors are small in relation to the cheek teeth.
There are two lower incisors large enough to be
referred to as castorid. UF175401 is larger and less de-
graded than UF175402. Both are smooth surfaced with
gently convex enamel faces (Fig.6:C). Three incisors are
referable as upper incisors to Castoridae based again on
size and degree of curvature. UF 163681 (Fig.6:Ca)
shows the occlusal surface. UF175404 and UF175403
are fragments of posterior portions. All three are simi-
lar, except for larger size, to the lowers with smooth an-
terior faces that are convex in cross section.
Several deciduous lower and upper premolars were
found. They resemble the corresponding P4 in general
morphology but are lower crowned, smaller, and have
less complex fossettids (Fig.6:D, E, L).
The p4 is similar to the lower molars except for its
larger size. It is also more elongate, and the parafossettid
is more quadrate. In all the lower cheek teeth the
hypoflexid is directed posterobuccally and nearly con-
nects with the metafossettid. The mesoflexid is the same
width as the metafossetid. The mesoflexid's connection
to the buccal margin is quickly lost with wear. The
parafossetid is an elongate "S" shape that connects with
the mesofossetid in the unworn condition. With wear,
the parafossettid becomes an isolated "L" shape located
toward the lingual margin. The m3 is the smallest of the
lower teeth and differs from the other teeth in that the
mesoflexid tends to completely divide the tooth into a
separate anterolophid and posterolophid.
The P4 is larger than the upper molars and simi-
lar in morphology except for having a less complex
parafossettid (Fig. 6:M, N). On all upper cheek teeth
a deep hypoflexus extends anterobuccally to meet the
shallow paraflexus. The mesoflexus runs from the
buccal margin lingually unto the midline where it turns
posteriorly to form an "L" shape in slightly to moder-
ately worn specimens. In more worn specimens the
posterior portion of the mesoflexus forms a separate
hypofossette. In slightly worn teeth a shallow com-
plex metaflexus lies on the posterobuccal corner. The
metaflexus is usually joined with a smaller fossette
that persists longer with wear than the metaflexus.
On the anteroloph of the MI and M2 a complex
parafossette is present that forms a "Y" shape in less
worn teeth and separates into three fossettes with con-
tinued wear. The M3 is significantly smaller, by 50-
60%, than P4. The posteroloph is compressed trans-
versely and the mesoflexus curves anteriorly, unlike
the other molars (Fig. 6:R).
DISCUSSION.-Although there are numerous

early Miocene beavers from the Great Plains (Stirton,
1935; Macdonald, 1963; Martin, 1987) few castorids
have been reported from Gulf or Atlantic Coastal Plain
regions before the late Barstovian (Hulbert, 1992) and
early Clarendonian (Webb et al., 1981). Wood and Wood
(1937) listed two teeth as Palaeocastor from the Texas
Oakville formation (early Miocene) and Albright (1996)
described the only significant Arikareean Gulf Coast
occurrence of a beaver from the Toledo Bend LF as
"Monosaulax" hesperus.
North American Arikareean Monosaulax was re-
viewed by Korth (1996) and placed into a new genus
Neatocastor. Neatocastor is separate from younger true
Monosaulax that is associated with later Miocene
"Merychippus faunas" (Stirton, 1935:416). Korth agreed
with Xu (1994) that the two Arikareean species of
Monosaulax, M hesperus and M. complexus, are syn-
onymous. However, the American species were not the
same as the Eurasian genus Stenofiber. Most recently,
Korth and Emry (1997) revised Anchitheriomys and
placed this genus along with Neatocastor and
Agnotocastor into a new subfamily, Agnotocastorinae.
Within this new subfamily, Agnotocastor of Whitneyan
age is the ancestral genus for Arikareean Neatocastor
followed by younger Anchitheriomys.
The Brooksville 2 sample, although similar in size
to Palaeocastor nebraskensis, is separable from the
Palaeocastorinae and assignable to the Agnotocastorinae
as indicated by the comparatively smaller convex inci-
sors and more quadrate cheek teeth. Palaeocastor tends
to have a shallow mesoflexus that is lost with wear mak-
ing the cheek teeth more oval shaped. Further, Martin's
(1987) partial diagnoses for the Palaeocastorinae, which
includes the genera Palaeocastor, Pseudocastor,
Fossorocastor, Euhapsis and Capacikala, is that the
incisors are enlarged and flattened and the fossettids are
relatively simple, whereas in the Agnotocastorinae the
fossettids are more complex and trend toward higher
complexity in younger species.
Korth and Emry's (1997) cladistic analysis defines
three plesiomorphic characters of Agnotocastor, reten-
tion of P3, relatively smaller crowned cheek teeth, and
less complex fossettids. The lower crown height and less
complex fossettids clearly show that the Brooksville 2
sample represents Agnotocastor. However, cranial char-
acters rather than dental traits define most genera and
species of fossil Castoridae. Therefore, no specific as-
signment is reached.
As discussed earlier, Albright (1996) referred the
only previously described Arikareean Gulf Coast
Castorid to "Monosaulax" (now Neatocastor)
hesperus. The Brooksville 2 species is significantly






Figure 6. Teeth ofAgnotocastor sp.: (A) Composite buccal view of Lp4-m3, x3; (B) Comparison of a) BKVL 2 LF Lp4 with
b) Toledo Bend LF Neatocastor hesperus Lp4, x3; (C) Cross Section of a) upper incisor and b) lower incisor, x3; D-R,
magnified x6; (D) UF163631 RDp4; (E) UF163622 LDp4; (F) UF163604 Rp4; (G) UF163606 Lp4; (H) UF163612 Lm2; (I)
UF163608 Rml; (J) UF163752 Lm3, unworn; (K) UF163619 Rm3; (L) UF163756 DP4; (M) UF163601 LP4; (N) UF163603
RP4; (0) UF163758 RMI or 2, unworn*(P) UF163615 LMI or 2; (Q) UF163616 RMI or 2; (R) UF163629 LM3.






HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hemando County, Florida

TABLE 5.-Dental measurements ofAgnotocastor sp. from Brooksville 2 LF.

incisor (UF175401) (UF175402)
ap 2.9 2.2
tr 2.6 2.4

dp4 (UF163631) (UF163622) (UF163626) (UF163625) N M OR S
ap 3.4 3.2 3.6 3.9 4 3.52 3.2-3.9 0.3
tr 2.6 2.2 2.4 2.4 4 2.4 2.2-2.6 0.16

p4 (UF163621) (UF163666) (UF163604) N M OR S
ap 3.6 4 3.7 3 3.77 3.6-4.0 0.21
tr 3.3 3.5 3.2 3 3.34 3.2-3.5 0.15

ml or 2 N M OR S
ap 10 3.03 2.8-3.3 0.21
tr 10 2.87 2.4-3.2 0.28

m3 (UF163752) (UF163619)
ap 2.6 2.7
tr 2.1 2.2

incisor (UF163681) (UF175404) (UF175403)
ap 3 3.3 3.3
tr 2.9 2.9 2.8

dP4 (UF163756)
ap 2.5
tr 3.2

P4 (UF163603) (UF163601) (UF163602) (UF163759) N M OR S
ap 4 3.6 3.9 3.6 4 3.76 3.6-4.0 0.21
tr 4.5 3.8 4.4 3.2 4 3.96 3.2-4.5 0.6

M1 or 2 N M OR S
ap 8 2.75 2.7-2.8 0.05
tr 8 2.81 2.6-3.0 0.15

ap 7 2.4 2.2-2.6 0.12
tr 7 2.31 2.2-2.6 0.09

lower crowned than the Toledo Bend sample (Fig.6:B)
and the fossettid morphology is not as complex. Fol-
lowing the relationship proposed by Korth between
Neatocastor and Agnotocastor, the Brooksville 2
sample lies near an intermediate stage between the

Whitneyan Agnotocastor and the late Arikareean
Hulbert's summary of Florida fossil vertebrates
(1992) listed an early late Arikareean castorid from the
Cow House Slough LF (UF40189) as indeterminate. This

new sample from Brooksville 2 shows clearly that this
specimen is also referable to Agnotocastor.
Later occurrences of castorids in Florida include two
genera, Anchitheriomys and Eucastor, found in later
Barstovian sediments of the Bone Valley (Hulbert, 1993)
and a species ofEucastor recovered from the Clarendonian
Love Bone Bed locality (Webb et al., 1981). Castorids
are notably absent from the Florida record during the late
Arikareean, the Hemingfordian, and into the early
Barstovian. This absence may be a sampling bias due to
the sparseness of samples in localities of these ages.

Family LEPORIDAE Gray, 1821
Subfamily PALAEOLAGINAE Dice, 1929
Genus MEGALAGUS Walker, 1931
Megalagus abaconis new species
Figures 7, 8, Table 6

ETYMOLOGY.-abaconis (Latin genitive):
named for K. Abaco Richardson.
HOLOTYPE.-UF 163640, Lp3
OCCURRENCE.-Brooksville 2 LF, Hernando
Co., Florida, "medial" Arikareean UBI
UF163577 LM1, UF163578 RM2, UF163579 LM2,
UF163580 RP4, UF163581 Rml, UF163582 Lp4,
UF163583 Rp3, UF163584 Lm2, UF163585 Lp3, UF
163586 RM?, UF163587 RM1, UF163588 M?,
UF163589 LP2, UF163590 Lp4, UF163591 Rm2,
UF163592 Lp4, UF163593 Lml, UF163594 Lm2,
UF163595 Rml, UF163596 Rp4, UF163597 LM?,
UF163598 Rm3, UF163599 RP3, UF163632 LII,
UF163633 RI1, UF163634 LII, UF163635 M?,
UF163636 LI2, UF163637 RMI, UF163638 RP4,
UF163639 Rdp4, UF163641 LP2, UF163642 Lp3,
UF163643 Rp3, UF163644 Rp3, UF163645 Rml,
UF163646 RdP3, UF163647 RM?, UF163648 M?,
UF163649 RP2, UF 163650 Ldp4, UF163651 Rdp3,
UF163652 Rdp4, UF163653 Lm3, UF163658 Ldp3,
UF163682 Lm3, UF163683 LM?, UF163684 dP?,
UF163685 LdP3, UF163686 DP?, UF163687 LP4,
UF163688 DP?, UF163689 Rp3, UF163690 RMI,
UF163691 LMI, UF163751 LM?, UF163766 M3,
UF163773 LI1.
DIAGNOSIS.-Advanced Megalagus, in which
buccal roots on upper molars are reduced and lost quickly
with wear. Resembles Megalagus dawsoni in greater
degree of hypsodonty and cement than older species of
Megalagus such as M. primitivus. Separable from M.
dawsoni by shorter length of upper molars in relation to


width, and reduction of buccal roots. The p3 has shal-
low lingual reentrant, lost with slight wear. External re-
entrant directed anterolingually. Lingual bridge on lower
molars forms very late in wear. Lower unworn to mod-
erately worn teeth have crenulated anterior talonid mar-
gins. The P2 with reduced buccal lobe. The P3 anteroloph
small with cresentric valley open at margin.
DESCRIPTION.-Several lower deciduous
premolars have been recovered (Fig.8:A, B, C, D, E).
Deciduous lower premolars of Palaeolaginae have never
been described in the literature, so the terminology ap-
plied in Fig. 7 is new. A deep transverse valley separates
the trigonid into an anteroloph and metaloph. The
anteroloph is formed from the paracone and a buccal
parastylid. On unworn to slightly worn teeth the
anteroloph is lower in height than the metaloph and equal
to the talonid (Fig. 8:E). The metaloph consists of the
metaconid and protoconid connected by a cristid. A trans-
verse valley separates the talonid and trigonid. A dis-
tinctive entoconid is directed posterolingually. A lingual
valley divides the entoconid from the mesoconid. The
hypoconid lies on the buccal extension of the talonid.
The anterior margin of the talonid is narrow and con-
cave. With wear the anterolophid and metalophid join
and a central enamel connection forms between the tal-
onid and trigonid (Fig.8:C).
The holotype of this new species is a single little-
worn p3 (Fig.8:F). The large cement-filled external re-
entrant has a crenulated posterior enamel surface. The
internal reentrant is shallow and disappears quickly with
wear. An anterior fold is present that extends minimally
down the tooth. In all the p3s the external reentrant is
curved anteriorly rather than extending lingually, as in
Palaeolagus philoi, and the anterolophid is directed
posterobuccally. With wear the p3 becomes larger in
occlusal outline (Fig.8:H, I).
The p4-m2 are all similar in size and morphology.
The differences among these teeth is the degree of an-
teroposterior curvature that increases in the posterior
teeth. The talonid and trigonid form two separate col-
umns. The trigonid is the larger of the two. In unworn
specimens a posterolophid is present on the talonid
(Fig.8:O, P). The trigonid and talonid are connected by
cement (Fig.8:P, Q, R) until very late wear stages when
a lingual enamel connection forms (Fig.8:P, T). The an-
terior enamel margin of the talonid is thin and crenu-
lated, whereas on the trigonid enamel is absent from the
anterior margin. The enamel is thicker on the buccal
margin and thins lingually.
The m3 is reduced in overall size with the talonid
reduced even more in relative proportions (Fig.8:P, Q).
The m3 is less hypsodont than the other molars.

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hemando County, Florida

- Metastyle
- Metacone
- Crescentric

A B C Valley
Trigonid: Anterolh Met ph alonid/Postero oph
Metaconid -- Mesoconid

Parastylid Hypoconid

Figure 7. Dental terminology used for Lagomorphs. Partially after A.E. Wood (1940) and Dawson (1958). (A) Lp3; (B)
slightly worn lower molar; (C) upper unworn molar; (D) decidious lower premolar.

TABLE 6.-Dental measurements for Megalagus abaconis n. sp. from Brooksville 2 LF.


2.47 2.3-2.7
1.54 1.5-1.7
1.71 1.5-1.9

2.23 1.9-2.4
2.12 1.8-2.5

2.6 2.5-2.7
2.62 2.5-2.9
2.15 2.0-2.2

2.46 2.4-2.6
2.46 1.6-2.7
1.94 1.6-2.1

2.63 2.6-2.7
2.6 2.6
2.23 2.2-2.3

1.58 1.1-1.9
1.48 1.3-1.7
1.1 .9-1.4






0.08 2.4
0.19 1.5
0.1 1.7

0.08 2.4
0.39 1.6
0.17 1.6

0.06 2.3
0 2.4
0.06 1.9













2.2-2.4 0.11
1.5-2.3 0.39






M3 (UF163766)
ap 1.1
tr 2





























HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hemando County, Florida

Both upper incisors have been recovered. There is
a single anterior sulcus that divides each incisor into two
lobes. The sulcus is located lingually off-center, so the
buccal lobe is the largest. The posterior incisor is simi-
lar to the anterior incisor except for smaller size and the
lack of an internal vascular canal.
The P2 has two anterior reentrants that divide the
tooth into three lobes. The buccal lobe is half the length
of the lingual lobe. The buccal reentrant is very shallow,
while the lingual reentrant extends almost to the crown
margin. There are two roots (Fig.8:T, U, V).
In P3 the anterior lobe is very narrow relative to the
posterior lobe. The cresentric valley is open to the ante-
rior margin and the buccal valley is moved inward lin-
gually. The P3 is single-rooted. The central valley is wide
and open. The central valley between the anterior and
posterior lobes of the upper premolars and first molar is
more prominent, while the hypostria is weak or absent.
M2 shows the reverse condition with a stronger hypostria
(Fig.8:AB) and less developed central valley.
Buccal roots are present in slightly worn molars
although they are small and disappear with later wear.
In the fourth premolar buccal roots appear to develop
sporadically or perhaps on more developed crowns
(Fig.8:W, X). In unworn molariform teeth the cresentric
valley margin is crenulated and a shallow depression lies
behind the metacone and lingual to a conical paracone
(Fig.8:W). As the molars wear, the buccal enamel is lost
and the lingual enamel at the transverse comers thick-
ens. The cresentric valley enamel wears first to a thick
area confined to the metacone region and then wears to a
uniform thickness.
The M3 resembles the P2 with two reentrants. How-
ever, in comparison with the P2, the reentrants are lo-
cated on the posterior margin, the lobes are of equal size,
and the tooth is higher crowned (Fig.8:A, E). The M3 is
reduced relative to the other molars. The posteroloph is
absent and the anteroloph has been compressed
DISCUSSION.-Lagomorphs have not been de-
scribed in detail from the Arikareean of the southeastern
United States or Florida. Morgan (1989) listed

?Palaeolagus in the White Springs LF. On this basis,
Hulbert (1992) listed the Palaeolaginae as ranging
through the late Oligocene into very early Miocene in
Florida. Like the beavers, rabbits are curiously absent
from the Florida record from the late Arikareean (c.21
Ma) until the late Barstovian (15 Ma) when Hypolagus
appears (Hulbert, 1992).
Elsewhere, there is a diversity of Arikareean lago-
morphs (Wilson, 1949; Dawson, 1967; Stevens et al.,
1969; Korth, 1992) with three genera of leporids,
Palaeolagus, Archaeolagus, and Megalagus, and two
genera of ochotonids, Gripholagomys and Desmato-
lagus. These genera are not all present at the same time
interval during the Arikareean. Palaeolagus disappears
after the early Arikareean to be replaced by its descen-
dant Archaeolagus (Tedford et al., 1996). Megalagus is
represented in the Arikareean by two species: M.
primitivus from the Harrison Formation of Wyoming
(Dawson, 1967) and M dawsoni, described by Black
(1961) from the Fort Logan Formation of Montana.
The Brooksville 2 rabbit is separable from the
Ochotonidae by the possession of lower and upper M3s.
In some morphological characters and size (Table 7) the
Brooksville 2 sample resembles the advanced
Palaeolagus philoi. Both have crenulated anterior mar-
gins on the talonid and the p3 produces a single reen-
trant with slight wear; this feature is also found in younger
Archaeolagines. However, all species of Megalagus pos-
sess these characters, which are probably convergent with
the Palaeolagus-Archaeolagus lineage. The retention of
buccal roots, primitive P3, and relation of valleys and
hypostriae on the upper molars serve to differentiate the
Brooksville 2 Megalagus from Palaeolagus and
Except for its smaller size, the p3 compares well
with those ofMegalagus brachyodon andM primitivus
depicted by Dawson, 1958:17, Fig.7a. Megalagus spe-
cies tend to have an external reentrant that extends
anterolingually and a posteroloph that is directed con-
versely to the reentrant. Palaeolagus and Archaeolagus
that have a single external reentrant tend to have it lin-
gually directed with a more oval shaped p3. A lower

Figure 8. Megalagus abaconis n. sp., x6 scale bar =1 mm. (A) UF163693 Rdp3 or 4; (B) UF163651 Ldp3 or 4; (C) UF163652
Ldp3 or 4; (D) UF175424 Rdp3 or 4; (E) UF163693 lingual view; (F) UF163640 Lp3, holotype; (G) UF163643 Rp3; (H)
UF163644 Rp3; (I) UF163642 Lp3; (J) UF163593 Lml; (K) UF163596 Rp4; (L) UF163519 Lm2; (M) UF163581 Rml; (N)
UF163582 Rml; (0) UF163582 Lp4; (P) UF163576 Rmn3; (Q) UF163598 Rm3; (R) UF163646 RdP4, occlusal view; (S)
UF163646 RdP4, anterior view; (T) UF163589 LP2; (U) UF163641 RP2, occlusal view; (V) UF163641 RP2, anterior view;
(W) UF163638 RP4?, unerupted; (X) UF163586 RMI?, unworn; (Y) UF163599 RP3; (Z) UF163589 RP4; (AA) UF163637
RP4; (AB) UF163751 RM2, occlusal view; (AC) UF163751 RM2, anterior view; (AD) UF163766 LM3, occlusal view; (AE)
UF163766 LM3, posterior view.

dentary, F:AM 99266, of "Megalagus sp." from the
White River Group, preserves dp3 and dp4 that are simi-
lar to those in the Brooksville 2 sample. The P3 is typi-
cal of Megalagus. In Palaeolagus the lobes of P3 are
nearly of equal width and the tooth is more molariform.
The morphology of the lingual portion of the upper mo-
lariform teeth is more indicative of Megalagus as well.
Palaeolagus tends to have weak lingual valleys and stron-
ger hypostriae on the anterior premolars and molars
whereas the opposite condition is present in Megalagus
(Wood, 1940).


With the above comparable characters the
Brooksville 2 leporid is assignable to Megalagus. This
brings the total to three species in the Arikareean. M
primitivus with its more primitive morphology begins
the early Arikareean and then the advanced lines of M
dawsoni in Montana and the new Brooksville 2 LF spe-
cies in Florida extend into the later Arikareean.
Cow House Slough LF produced a number oflago-
morph teeth (UF40155-61, 40190) that I am referring to
Megalagus abaconis because of their corresponding
morphology and size. One upper molar (UF40155) has

TABLE 7.-Comparative dental measurements for Whitneyan-Arikareean Leporidae. Compiled from Dawson 1958,
1967; Black 1961; Stevens et al. 1969.

Megalagus M. M. M. Palaeolagus P Archaeolagus A.
abaconis dawsoni cf.primitivus turgidus hypsodus philoi acricolus ennisianus

p3 ap


p4 ap 2.5-2.7
tra 2.5-2.9
trp 2.0-2.2

ml ap

for lower


m2 ap 2.6-2.7
tra 2.6
trp 2.2-2.3

m3 ap


P2 ap 1.1-1.4
tr 1.9-2.1

P3 ap

P4 ap 2.1
tra 3.4

Ml/ap 1.8-2.1
M2 tra 2.7-3.6

M3 ap 1.1
tr 2







1.76 2.0-2.2
2.14 2

3.28 2
2.4 2.1-2.2
2.1 1.6-1.8

3.42 2.1-2.3
2.8 2.1-2.4
2.45 1.9-2.0

3.27 2.1
2.74 2.1-2.3
2.25 1.7-1.8





x x
x x







1.9-2.5 1.7-2.1
1.9-2.2 1.7-2.0













2 1.6-1.7
4.0-4.2 3.5-3.6

1.9-2.1 1.6-1.7
3.4-3.9 2.9-3.3

x 0.6-0.7
x 1.3











HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hlernando County, Florida

an unusual development of buccal roots that extend not
from the lower margin but from two different heights
starting halfway up the crown.
The single preserved tooth of a lagomorph from the
White Springs LF (UF121427) has a well-developed
hypostria and cresentric valley like the molars of
Megalagus abaconis. On the other hand, the tooth is
considerably smaller (ap 1.2; tr 2.7) than specimens from
Brooksville and Cow House Slough, which suggests that
the White Springs rabbit is a different taxon, possibly
assignable to Archaeolagus.

Order CARNIVORA Bowdich, 1821
CARNIVORA incertae sedis
Genus PALAEOGALE von Meyer, 1846

COMMENT.-Palaeogale is a carnivore of un-
certain affinities that has been allied at different times to
Mustelidae (Simpson, 1946) and to Feliformia by de-
scent from early Tertiary Viverravidae (Flynn and
Galiano, 1982; Hunt, 1989). The confusion over its taxo-
nomic position is because Palaeogale has both mustelid
and feliform dental features. Among mustelid features is
the reduced m2, while feliform characters include a slit-
like camassial notch, loss of metaconid on m 1, and pres-
ence of a parastyle on P4. Flynn et al. (1988) proposed
that Palaeogale be placed in Carnivora incertae sedis
due to its lack of clear synapomorphies with either cani-
forms or feliforms and their classification is followed

Palaeogale minute
Figures 9, 10, Table 8

TYPE LOCALITY.-St. Gerand le Puy, France
RANGE.- late Oligocene-early Miocene
UF163537 LP4 w/maxilla- alveoli for Ml, UF163539
RP3 w/maxilla, UF163540 LP4, UF163544 LP4,
UF1636545 RP4, UF163546 Lml, UF163548 L. man-
dible w/ml-m2, UF163549 R. mandible w/p3-ml -alveoli
for m2, UF163550 L. mandible w/p3-ml and alveoli for
i3,c,p2-p3, UF163692 LP4, UF163718 Lp4, UF163749
L. edent. mandible-alveoli for p3-m2
DESCRIPTION.-The pi is absent (UF 163550).
The p2 is double rooted alveolii in UF163550). The p3
is transversely compressed and elongate, as are the rest
of the lower teeth. The p4 has a posterior accessory cusp
and a cingular cusp on the posterior margin. The lower
camassial has lost the metaconid, the talonid is reduced

and unbasined with the cristid obliqua forming a central
longitudinal shearing crest that ends posteriorly in a small
cusp (hypoconid?), the protoconid is high thin and sharp,
the paraconid is positioned anteriorly, thin, and blade-
like. A cingulid is not present. UF163548 preserves the
only m2. It is double rooted, small, and oval in outline
with a low central ridge that is slightly pinched at the
middle of the tooth.
The P3 is the first upper premolar in the Brooksville
2 LF sample. It has a small posterior heel and no acces-
sory cusps. The upper camassial has a deep and slit-like
notch separating the strong metacone blade from the pos-
teriorly slanted paracone. The protocone of the P4 is di-
rected anteriorly and falls in line with a sharp parastyle
located on the anterolingual corer. A "U-"shaped notch
separates the parastyle and protocone. The P4 has a slight
lingual cingulum that begins slightly anterior to the pos-
terior margin of the metacone and ends at the camassial
notch. Only the alveoli represent Ml. The Ml would
have been strongly compressed transversely and was even
more reduced relative to the size of the P4 than in the
M 1 of the older Palaeogale sample from the White River
DISCUSSION.-Palaeogale is unique in combin-
ing the size of a small mustelid with a hypercamivorous
dentition featuring the lack of a metaconid on the ml,
transverse compression of all teeth, and reduction of M2.
The Brooksville 2 sample exhibits all the characteristic
features of Palaeogale dental morphologies (Simpson,
1946) and is similar to specimens of Palaeogale housed
in the American Museum.
De Bonis (1981) reviewed all Eurasian and North
American Palaeogale. He synonymized Palaeogale spe-
cies into a total of four taxa. The taxonomy was based
on size difference, presence of M2, loss of pi, and age
of occurrence. He argued that Palaeogale was conser-
vative morphologically, had a large size range within
species, and that there were insufficient characters used
to separate most described species. Simpson (1946) ear-
lier suggested that two species of Palaeogale from the
Oligocene of Mongolia, P. ulysses and P parvula de-
scribed by Matthew and Granger (1924), might be syn-
onymized in the belief that the smaller group were fe-
males and the larger were males, as in many modem
The four species of de Bonis were diagnosed as fol-
lows: (1) Palaeogale sectoria encompasses all known
late Eocene and early Oligocene species, is medium sized,
and retains M2 and p1; (2) P. minute, the smallest spe-
cies, is found in the late Oligocene and early Miocene of
Europe and North America, has lost M2 and pi, and,
while relatively small, has a very large intraspecific size




Figure 9. Palaeogale minute, x3, scale bar =1mm. (A) UF163550, buccal view; (B) UF163550, lingual view; (C) UF163548,
buccal view; (D) UF163548, occlusal view; (E) UF163537, lingual view; (F) UF163537, occlusal view.

range with P4s that are 6.5 mm to 3.5mm; (3) P.
hyaenoides, found only in the Miocene of Europe and
medium sized; and (4) P dorothiae, described by
Macdonald (1963) in the Arikareean age Wounded Knee
fauna of the Great Plains in North America, is also the
largest of the Palaeogale species.

The Brooksville 2 Palaeogale is smaller than most
other samples of P. minute. The largest specimen,
UF163544, a RP4 with a length of 4.6mm, falls in the
lower range ofP minute (see Fig. 10) whereas the small-
est, UF163540, also a RP4 with ap=3.25, falls below
the range by several tenths of millimeters. Because of

TABLE 8.-Dental Measurements ofPalaeogale minute from Brooksville 2 LF

(UF 163550) (UF 163549) (UF 163548)











(UF 163536)
P3 ap 2.96
tr 1.59


P4 ap 7
tra 7
trp 7





93-1 24

0 14

p2 alv

p3 ap

p4 ap

ml ap

m2 ap

ml ap

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hemando County, Florida

the extensive size range of P minute, as described by de
Bonis, I am referring the Brooksville 2 sample to P.
minute. It is possible that the Brooksville 2 Palaeogale
is an even smaller species that is more hypercamivorous.
Recovery of the MI might show that the Brooksville 2
sample has a greater reduction of this tooth compared to
other species of Palaeogale.
Albright (1996) described a partial P4 (LSUMG V-
2246) from the Toledo Bend LF as ?Palaeogale sp. His
referral to Palaeogale centered on the presence of a car-
nassial notch. In modem mustelids the camassial notch is
lost and the metacone blade is not separated from the para-
cone. This tooth differs from Palaeogale, however, and
may best be referred to the family Mustelidae, for the rea-
sons below. While eumustelids have lost the camassial notch,
mustelids that are known from the Oligocene to early Mi-
ocene, often informally called "palaeomustelids" (Baskin,
1998b), tend to retain the notch but develop mustelid fea-
tures in the upper and lower molars. As Albright noted, the
tooth is most nearly comparable to specimens in the Frick
collection of the American Museum labeled "Miomustela"
that retain a metaconid on the lower camassial and thus are
closer to modern mustelids. Finally, the Toledo Bend speci-

3 w




1 ~

[ Palaeogalei
A Palaeogalej
0 Palaeogale

men has an extensive lingual cingulum similar to mustelids,
whereas Palaeogale has only a slight cingulum under the
lingual metacone.
The Brooksville 2 sample marks the first substan-
tial occurrence of Palaeogale in the Gulf Coastal Plain.
De Bonis (1981) listed a few small, undescribed
Hemingfordian specimens from the plains region of North
America as P minute. These specimens are probably
congeneric with the Brooksville 2 taxa although they
are slightly larger. He found no records or specimens
other than P dorothiae (Macdonald, 1963) that are Late
Oligocene in age. This author observed a ml trigonid of
Palaeogale, AMNH 81039, from the lower Harrison
formation (early late Arikareean? ofTedford et al., 1987),
that is comparable to the Brooksville 2 Palaeogale. There
are also undescribed specimens of Palaeogale in the
McCann Canyon LF, placed by Korth (1992) in the
"early late" Arikareean.
In addition to the Brooksville 2 sample, there are a
few specimens of Palaeogale minute from SB-lA LF
They include UF117589, an undescribed right maxilla
fragment with P3-P4 and ml alveoli, and UF163888, a
partial LP4.



4 5 6 7 mm
minute (B2LF)
minute (type)
postelina (type)

Figure 10. Distribution of ml as a function of length x width for late Oligocene to Miocene Palaeogale. Open Circles are
Eurasian samples, open box is P. minuta-holotype, closed triangle is P "postfelina "-holotype. The Brooksville 2 LF Palaeogale
is represented by diamonds.

Suborder CANIFORMIA Kretzoi, 1943
Infraorder ARCTOIDEA Flower, 1869
Parvorder MUSTELIDA Tedford, 1976
Superfamily MUSTELOIDEA Fischer de Waldheim,

COMMENT.-The record of early musteloids
from the Paleogene and earliest Miocene of Europe is
quite rich (Helbing, 1930, 1936; Dehm, 1950; Ginsburg,
1961; Lange, 1970; de Bonis and Guinot, 1987; Ginsburg
and Morales, 1992; Wolsan, 1993; Cirot and Wolsan,
1994). In contrast, the North American record is sparse.
Clark (1936) described the only early North American
musteloid, Mustelavus priscus, from the late Eocene
Chadron Formation on the basis of a single skull and
jaws. The next record in North America is in the "early"
Arikareean when "?Plesictis" sp. is reported from the
Monroe Creek Fauna (Macdonald, 1970) and Promartes
appears (Tedford et al., 1987). In the later Arikareean
Oligobunis appears in North America and throughout
the rest of the Tertiary Old World musteloid immigrants
continue to invade North America and add to the diver-
sity of the already present taxa (Baskin, 1998b).
The Brooksville 2 musteloids, described below, in-
dicate that the early musteloid diversity in North America
is greater than previously known. The Brooksville 2
musteloids may be part of an unknown autochthonous
North American radiation derived from Mustelavus or
they may represent immigrants from Asia that differ from
the European musteloids. Unfortunately, the Oligocene
record of musteloids in Asia is even less well known
than that of North America (B. Wang, 1992), so explo-
ration of the latter hypothesis is not possible. Compari-
son of the Brooksville 2 material with other Florida lo-
calities, SB-lA LF and 1-75 LF, and the Toledo Bend
LF (see Palaeogale discussion) has revealed other early
musteloids. These specimens help close the gap between
the late Eocene occurrence ofMustelavus and the record
of Promartes and "?Plesictis" in the late Oligocene.
The radiation of small arctomorph carnivores dur-
ing the Oligocene and early Miocene may include taxa
that are convergent with true mustelids and procyonids
but represent non-mustelid and non-procyonid lineages
as a part of an undifferentiated stem group whose rela-
tionships are controversial (Schmidt-Kittler, 1981;
Wolsan, 1993; Wolsan and Lange-Badre, 1996). Alter-
natively, Baskin (1998a, 1998b) argued that most of the
primitive musteloid taxa could be placed in either
Procyonidae or Mustelidae on the basis of dental char-
acters. Any musteloids that have a reduced m2/M2 along


with a reduced postprotocrista and metaconule are con-
sidered to be mustelids. However, Hunt and Tedford
(1993) suggested that features of the auditory region are
more reliable for resolving such complex phylogenetic
relationships. If there were several radiations of
musteloid groups, then dental features may be subject
to convergence and may yield misleading phylogenetic
Since no specimens of the Brooksville 2 musteloids
preserve the auditory region the relationships of these
taxa to mustelids and procyonids is uncertain. There-
fore, I do not assign them to either family but recognize
that they are part of the early diversity of musteloids.
Because of the rarity of early North American
musteloids, I have used Wolsan's (1993) character analy-
sis of early European musteloids as the framework for
discussion and comparison in the systematics. Table 9 is
a modified character matrix from Wolsan (1993:350,
Table 1) to which I have added two new genera from the
Brooksville 2 taxa, Promartes, and two more dental fea-

Acheronictis webbi new genus and species
Figurell, Tables 9, 10, 11

ETYMOLOGY.-Genus, Acheron (Gr.) river of
the netherworld; iktis (Gr.) weasel: species, named for
S. David Webb
OCCURRENCE.-Brooksville 2 LF, Hernando
Co., Florida, "medial" Arikareean UBI. ?Monroe Creek
F "medial" Arikareean UBI.
maxilla frag., UF163538 L. dentary with p2, alveoli for
c,pl, UF163547 RP4, UF163703 trigonid Lml,
UF163746 RMI, UF163748 L. dentary with p2, alveoli
for c, p1, p3-p4
DIAGNOSIS.-Acheronictis is 10-20% smaller
than Mustelavus and Angustictis (Table 11). Acheronictis
shares the most dental characters with Mustelavus and
Mustelictis (Table 9). Can be separated fromMustelavus
by the following dental characters: postprotocrista ab-
sent; Ml more quadrate (stylar region and anterolingual
cingulum wider; transverse width less); Ml protocone
lies closer to anterior margin; Ml metaconule variably
present (when present it is more distinct than in
Mustelavus); protocone of P4 less distinct; cingular heel
on P3 larger; buccal cingulum on P3 absent; and prin-
ciple cusps of lower premolars distinguished from
Mustelictis by smaller size (Mustelictis is 30-50% larger).

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

TABLE 9.-Distribution of the states of 17 dental characters in 12 genera of North American and European mustelidans, new
Brooksville 2 LF taxa in boldface. The character states are defined below and follow those of Wolsan (1993: Figs. 4-6,
characters 12, 14-27), except for 1 and 2 which are added by this author. An "a" designates the primitive state for each char-
acter; "b, c, d, and e" indicate derived states. Polarity of states as defined by Wolsan. In instances in which two different states
of the same character could be scored for each genus, both states are listed in the matrix according to the assumed sequence in
which they appeared in each genus. Missing data denoted by "?". Characters in brackets are based on extrapolated data. First
numbers in parentheses after taxa are characters in common with Arikarictis n. gen.; second number italicized in parentheses
is characters in common with Acheronictis n. gen. All the taxa listed below occur in the late Oligocene and early Miocene,
possess a carnassial notch on P4 (Plesiogale and Paragale have lost this feature), and lack a hypocone on P4.

Taxon Characters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Stromeriella (5),(8) a a b a a a b b a a b a a a a a a
Franconictis (8), (8) a a [b] b a a a b a c b a a a a a b
Amphictis (6), (8) a a ab a a a a b a ac b a a a a a a
Mustelictis (9), (10) a b b b a a a ab a ac b a a ab a a b
Plesictis (7), (5) c c b b a a b b a de b ab b b ab a b
Psuedobassiris (7), (9) b b b b a a a a a ac b a a b a a ab
Broiliana (4), (6) a a a ab a a a b a a ab a a c a a a
Angustictis (7), (7) b a b b a a a b a c ab a a c a a a
Mustalavus (6), (11) a b b a a a a ab a ac b a a a a a a
Acheronictis c b b a a a a ab a ? b a a ? ? ? ?
Arikarictis b c [b] b a a a ab a c [b] ? ? ? ? ? ?
Promartes b c b b a a ab b a c b a b b a a b

Definitions of the states of characters: 1 a-postprotocrista present on MI, 1 b-postprotocrista weak on MI, 1 c-
postporotocrista absent on MI; 2 a-MI subequal width metastyle and parastyle region, parastyle only slightly raised, 2 b-
MI metastyle region reduced, height of parastyle less than paracone equal to metacone, 2 c-M1 metastyle region very
reduced, height of parastyle equal to or greater than paracone; 3a-PI two-rooted, 3b-PI single-rooted, 3c-PI absent; 4a-
P4 protocone not differentiated or crescentric, formed by cingulum entirely, 4b-P4 protocone conical, not formed by cingulum
entirely; 5a-P4 hypocone notably smaller than protocone or not differentiated, 5b-P4 hypocone and protocone subequal in
size; 6a-M1 larger than or equal in size to P4, 6b-MI smaller than P4; 7a-lingual half of Ml crown shorter than buccal
half, anterior and posterior borders of lingual half not parallel to each other, 7b-lingual half of Ml crown shorter than buccal
half, anterior and posterior borders of lingual half parallel to each other, 7c-lingual half of M1 crown about equal in length
to or longer than buccal half, halves separated from each other by anteroposterior constriction, 7d-lingual half of Ml crown
subequal in length to buccal half, no anteroposterior constriction between both halves; 8a-anterior and posterior cingula of
MI not continuous around lingual base of protocone, 8b-anterior and posterior cingula of Ml continuous around lingual
base of protocone; 9a-buccal border of M2 crown situated behind buccal half ofMl, 9b-buccal border of M2 crown situated
behind lingual half of Ml; 10a-M2 three-rooted and distinctly smaller than P4, 10b-M2 three-rooted and subequal in size
to P4 (Procyonidae), 10c-M2 two-rooted, 10d-M2 single-rooted, 10e-M2 absent; lla-pl two-rooted, llb-pl single-
rooted, lc--pl absent; 12a-ml trigonid less than three times as long as talonid, 12b-ml trigonid more than three times as
long as talonid; 13a-ml metaconid distinctly higher than paraconid, 13b-ml metaconid subequal in height to paraconid,
13c-ml metaconid distinctly lower than paraconid, 13d-ml metaconid not differentiated; 14a-ml entoconid and entoconulid
poorly differentiated, anterior and posterior halves of lingual wall of m talonid subequal to each other, 14b-ml entoconid
and entoconulid poorly differentiated or not differentiated, anterior half of lingual wall of ml talonid distinctly lower than
posterior half, 14c-ml entoconid prominent, ml entoconulid poorly differentiated or not differentiated, 14d-entoconid and
entoconulid prominent; 15a-m2 two-rooted, 15b-m2 single-rooted; 16a-m2 metaconid present, 16b-m2 metaconid not
differentiated; 17a-talonid basin of m2 distinctly longer than trigonid basin, 17b-talonid and trigonid basins ofm2 subequal
in length.


A ~





Figure 11. Acheronictis webbi n. gen and sp., scale bar =1 mm. (A) UF163536, LP3, occlusal view; (B) UF163536, LP3,
buccal view; (C) UF163547, RP4, buccal view; (D) UF163547, occlusal view; (E) UF163547, lingual view; (F) UF163745,
RMI, holotype, occlusal view; (G) UF163745, anterior view; (H) UF163745, posterior view; (I) UF163745, buccal view; (J)
UF163746, RMI, occlusal view; (K) UF163746, posterior view; (L) UF163748, L. dentary w/p2, buccal and occlusal views;
(M) UF163538, Lp2, buccal view; (N) UF163703, trigonid Lml, buccal view; (0) UF163703, occlusal view; (P) UF163703,
lingual view.

Mustelictis has a much wider M in relation to its length.
The Ml ofAcheronictis does not have a postprotocrista
unlike Mustelictis, Plesictis, Franconictis, and
Amphictis. Separable from Angustictis by the follow-
ing: MI is buccally as long but shorter in width by 20-
30%, P3 is smaller in length but broader in width (P3
1/w ratio similar to Mustelavus and Pseudobassaris).
Plesictis has a more derived stylar region on Ml, the
metastyle region is more reduced, and the height of the
parastyle is greater than the paracone. Franconictis and
Amphictis have a more primitive stylar region, the Ml
has a subequal metastyle and parastyle region, and the
parastyle is only slightly raised. In Plesictis the m meta-
conid is subequal in height to paraconid, while
Acheronictis has the ml metaconid higher than paraconid,

similar to Mustelavus, Mustelictis, and Amphictis. P3
similar length to width ratio as Mustelavus and
Pseudobassaris but smaller and broader than in
DESCRIPTION.-The small protocone on the P4
is anterolingually placed and forms a small subconical
heel rather than a distinct cusp (4a). The parastyle is
low and elongate as in typical mustelids. A weak lingual.
cingulum is present and extends the length of the meta-
cone. A weak buccal cingulum was probably present,
but wear has made this feature indeterminate. The car-
nassial notch is present.
The M is the most diagnostic of the recovered teeth
assigned to this species. The Ml is subtriangular in out-
line with a broad stylar shelf. The metacone, paracone,


HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

TABLE 10.-Dental measurements forAcheronictis webbi n. gen. et sp. (upper and lower teeth) andArikarictis chapini n. gen.
and sp. (upper teeth).

Acheronictis Arikarictis Acheronictis
webbi chapini webbi

(UF163536) (UF163538)
P3 ap 2.9 x p2 ap 2.6
tr 1.5 x tr 1.1
(UF163547) (UF163695) (UF163748)
P4 ap 4.5 7.4 p3 alv 2.7
tr 2.2 4.3
(UF163746) (UF163702) (UF163745) trigonid (UF163703)
M1 ap 3.1 5.5 3.6 (ml) ap 2.9
tr 3.7 8.1 4.2 tr 2
M2 ap x 1.2
tr x 4.1

and protocone are subequal in size. The paracone is the
tallest of the principal cusps. The parastyle is more de-
veloped than the metastylar region (2b). It is weakly con-
nected to the preparacrista and as tall as the metacone.
In UF163745 the preprotocrista forms a small distinct
cusp before connecting with the paracone (Fig. 11: G).
A minute paraconule is distinguishable in the
preprotocrista as a slight thickening and notch along the
crest. The postprotocrista is entirely absent (Ic). The
presence of the metaconule is variable: on UF163745 it
is absent, while UF 163746 has a small conical metaconule

toward the posterior margin. The lingual cingulum is nar-
row and thickens slightly posteriorly. It is continuous with
the posterior cingulum but ends above the buccal proto-
cone margin (8ab). Three roots are present. The lingual
is the largest and broad transversely. The buccal roots lie
on the buccal margin with the paraconal root larger than
the metaconal root. No M2 of this taxon was recovered.
Several lower dentary fragments can be allocated
to this genus. The only complete lower tooth that has
been recovered is p2, but there is also a trigonid of ml,
UF163703. The dentary fragments show that a single

TABLE 11.-Dental measurements forAcheronictis webbi n. gen. et sp. andArikarictis chapini n. gen. et sp. from Brooksville
2 LF and other Oligocene-early Miocene musteloids from North America and Europe. Comparative taxa are as follows: A,
Mustelavuspriscus; B Mustelavus sp.; C, Angustictis; D, Pseudobassaris; E, Mustelictis; F Broiliana; G, Plesictisgenettoides;
H, Amphictis milloquensis; I, Promartes.

Achero- Arika-
nictis rictis A B C D E F G H I
P3 ap 2.9 x 3.7 x 3.2 4.5-5.1 4.1 3.7 x x x
tr 1.5 x 1.9 x 1.2 2.5-3.3 2.3 1.8 x x x

P4 ap 4.5 7.4 5.2 7.6 4.8-6.0 5.8-7.2 6.5 5.9 7.4 8.5 9.4-10.6
tr 2.2 4.3 3.8 4.8 2.7-3.4 3.8-5.0 3.8 3.6 4.6 4.6 6.0-6.5

M1 ap 3.1-3.6 5.5 2.9 5.5 3.6-3.7 4.8-6.2 4.2 3.6 4.8 6.2 6.0-6.7
tr 3.7-4.2 8.1 5.3 7.2 5.2-5.7 6.7-7.8 7.4 6.0 7.2 10 9.8-10.0

M2 ap x 1.2 x x 1.8 3.0 x x x x 2.1-2.2
tr x 4.1 x x 3.9 5.7 3.6 x x x 5.1-5.6

rooted pi was present. The p2, UF163538 (Fig.12: M),
is low cusped, posteriorly wider, with the principal cus-
pid over the anterior root, and a cingular heel over the
posterior root. The p2 has weak lingual and anterior cin-
gula. Alveoli for p3 and p4 show that these teeth would
be progressively larger and show the same tendency to
be wider posteriorly. The m trigonid, UF163703 (Fig.12:
N, O, P), has a tall protoconid followed in height by the
metaconid (13a) and then the paraconid. The apices of
the cusps lie at the margins forming a lingually open
trigonid basin. The paraconid blade forms a deep car-
nassial notch with the protoconid blade. A small notch is
present in the valley between the protoconid and the meta-
conid. A short buccal cingulid runs from just posterior
to the paraconid to the anterior base of the protoconid.
DISCUSSION.-The material assignable to
Acheronictis is the smallest of any of the early musteloid
taxa. It is the same size as Palaeogale minute found at
Brooksville 2 (Table 8).
In dental characters Acheronictis is closest to
Mustelavus priscus and the European genus Mustelictis
(Table 9). This similarity may stem from the fact that
dentally Acheronictis is not far derived from the
plesiomorphic arctoid condition found in Mustelavus and
Derived features of Acheronictis include the loss
of the postprotocrista and enlarged parastyle region, traits
shared with many of the advanced late Oligocene Euro-
pean mustelids (Baskin, 1998b). Unfortunately, as dis-
cussed under Comment, there is no consensus on musteloid
relationships, and convergence of dental traits in Ache-
ronictis with the European genera cannot be ruled out.
Acheronictis could be equivalent to "?Plesictis" sp.
(Baskin, 1998b) from the Monroe Creek Fauna (SDSM
6264). They are close in size and morphology. However,
since "?Plesictis" sp. is based on a single fragment of
jaw, a sufficient comparison cannot be made to propose
formal synonymy.

Arikarictis chapini new genus and species
Figure 12, Tables 9,10,11

EYTMOLOGY.-Genus, Arikareean NALMA,
iktis (Gr.) weasel: species, named for Anne Chapin
OCCURRENCE.-Brooksville 2 LF, Hernando
Co., and SB-1A LF, Suwannee Co., Florida; "medial"
Arikareean UBI.
UF163699 LM2, UF175485 RP4; from Brooksville 2
LF. UF163782 RM2; from SB-1A LF.


Smaller than Promartes by 20-30%. Larger than
Mustelavus by 30-35%. Mustelictis is 10-20% smaller.
Similar in size to Pseudobassaris and Plesictis
genettoides (Table 11). More derived than Mustelavus
and Mustelictis in reduction of postprotocrista and en-
largement of parastyle on Ml. Mustelictis has a more
anteroposteriorly compressed M Angustictis differs by
20-30% larger size of P4 and M and 40% lower length
to width ratio of M2. On the M1 ofAngustictis the ante-
rior and posterior cingula are strongly continuous,
whereas they are only weakly connected in Arikarictis.
Arikarictis can be distinguished from Pseudobassaris
by the following dental characters: buccal margin of Ml
smaller in comparison to width, M hypocone more dis-
tinct, M posterior margin straight (not recurved towards
the anterior), Ml metastyle reduced, Ml parastyle con-
nected to paracone, P4 protocone less robust, and P4 not
as elongate in relation to width. Differs from Plesictis in
having convergent anterior and posterior margins of the
lingual half of Ml and two-rooted M2. Arikarictis dif-
fers from Broiliana, Franconictis, and Stromeriella by
more derived postprotocrista, parastyle, and Ml with a
more triangular occlusal outline. Promartes can be sepa-
rated from Arikarictis by: Ml parastyle not as well-de-
veloped into a distinct cusp (higher than paracone), MI
metastyle not as reduced, M anterior cingulum weaker,
and anterior and posterior margins of lingual half of M I
not parallel (in Promartes they are subparallel).
DESCRIPTION.-The teeth that have been recov-
ered from the Brooksville 2 LF are the upper carnassial
and molars. P4 retains the carnassial notch. The lingual
cingulum begins at the posterior margin of the metacone
and ends at the carnassial notch. A ridge runs from the
apex of the paracone to a short parastyle close to the
paracone margin. The protocone is rounded and anterior
of a position directly lingual to the paracone. A buccal
cingulum is present along the entire buccal margin but
thins along the margin of the paracone.
The Ml has a prominent parastylar region and sty-
lar shelf. The parastyle is as tall as the metacone and
formed from the intersection of the anterior cingulum,
the buccal cingulum, and the premetacrista. Along the
medial to anterior portion of the stylar shelf the buccal
cingulum consists of a series of small cuspules. The para-
cone is the tallest of the principal cusps followed by the
metacone and then the protocone. The cristae of the trans-
versely compressed metacone and paracone form sharp
ridges along the crests of these cusps. The trigon basin
is wide and shallow. The paraconule is a slight conical
thickening in the preprotocrista and is separated from
the protocone by a small notch. The preprotocrista joins

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida


Figure 12. Arikarictis chapini n. gen. and sp. UF163695 LP4, UF163702 LMI, UF163699 LM2, scale bar=2 mm. (A)
composite occlusal view; (B) composite buccal view.

the more buccal anterior cingulum in front of the meta-
cone. After the paraconule it extends slightly up the slope
of the protocone. The postprotocrista is a reduced ridge
only at the base of the protocone and at the margin of the
posterior portion of the lingual cingulum. The lingual
cingulum begins as a minute marginal anterior cingulum
above the notch of the paraconule. It thickens and ex-
tends in height directly lingual of the protocone, then
curves around posteriorly to join with the small but dis-
tinct metaconule, and then ends at the base of the meta-
cone. A series of minute cuspules give the posterior por-
tion the appearance of serrations. The lingual cingulum
is not continuous with the buccal cingulum.
The M2 is double rooted, half as wide transversely
as the M1, and compressed anteroposteriorly compared
to the M1. The mid-stylar region is prominent. The para-
cone is equal in size to the protocone. The metacone is
reduced and separated from the paracone by a small cen-
tral notch. The trigon basin is rounded, depressed up-
ward, and surrounded by a pre and postprotocrista that
extend from the centrally located protocone. The lingual
cingulum is evenly spaced around the protocone.
DISCUSSION.- Musteloids of this size and mor-
phology have not been described from the Arikareean of
North America. Comparison of this taxon with casts of
the European musteloids and comparison conducted by
Wolsan (written communication, 1998) with the original
European specimens shows that the Brooksville 2 mate-
rial represents a distinct genus.
Table 9 shows that Arikarictis resembles many of
the European musteloids including Mustelictis,
Angustictis, Pseudobassaris, and Plesictis. In overall
morphology the teeth are closest to Angustictis and

Pseudobassaris. These taxa have developed parastylar
regions and similar length to width ratios of the Mis.
The M2s are also comparably reduced (Table 11). How-
ever, as discussed previously, there is disagreement as to
the taxonomic placement of the primitive Musteloids.
Several species of Plesictis and Pseudobassaris have
been placed in Mustelictis (Schmidt-Kittler, 1981) or
alternately, Pseudobassaris and Angustictis have been
included in the Procyonidae, as its most primitive mem-
bers (Wolsan, 1993). Therefore, determination of the
precise relationships between Arikarictis and other primi-
tive musteloids is not possible at this time.

Infraorder CYNOIDEA Flower, 1869
Family CANIDAE Gray, 1821
Subfamily HESPEROCYONINAE L.D. Martin, 1989
Genus ENHYDROCYON Cope, 1879
Enhydrocyon cf. pahinsintewakpa
Figure 13

TYPE.-SDSM 53325
TYPE LOCALITY.-SDSM V5361, Sharps For-
OCCURRENCE.-Early to early Late
Arikareean; Lower Arikaree Group, Great Plains; "me-
dial" Arikareean UBI, Florida.
DESCRIPTION.-The M2 has two large roots
above the protocone and paracone and a reduced third
root above the metacone. The paracone is two-thirds
larger than the metacone. The metastylar region is very
diminished compared to the parastylar shelf. A thick lin-


Figure 13. Enhydrocyon cf. pahinsintewakpa. UF163523
LM2, scale bar =1 mm. (A) occlussal view; (B) posterior


gual cingulum evenly surrounds the trigon basin. The
protocone is worn away, its presence indicated by a shal-
low groove on the posterolingual corer. An anterior cin-
gulum is continuous with the lingual cingulum, thickens
in the parastylar region, and ends buccally beside the pre-
centrocrista. The tooth is 4.9mm long and 8.6mm wide.
DISCUSSION.-X. Wang (1994) reviewed the
Hesperocyoninae and in his diagnosis of Enhydrocyon
one of the defining characters is that the lingual cingu-
lum encircles the protocone. Other Hesperocyonines tend
to lose all traces of the anterior lingual cingulum on the
M2. Mesocyon and Cynodesmus have a lingual cingu-
lum that is thicker posteriorly and very thin anteriorly.
Furthermore, Cynodesmus tends to have a broadened
buccal region on the M2 with equal sized paracone and
metacone. In size the M2 is smaller than that of
Osbornodon the other medium sized canid present in the
Brooksville 2 LF. Comparison with specimens in the
Frick collection supports a close alliance to the
Enhydrocyon-Philotrox clade.
The small metacone and reduced metastylar region
are plesiomorphic features found in the older species of
Enhydrocyon and the sister genus Philotrox. The more
derived younger species, E. stenocephalus, E. basilatus
and E. crassidens, tend to have more equal lingual cin-
gulum and buccal regions. The relative size of the M2
also indicates a primitive form. Enhydrocyon basilatus
and E. crassidens, the terminal members of the group,
both have comparatively reduced M2s that are single
rooted or in some cases absent (X. Wang, 1994).
Sunkahetanka and Philotrox have the same size M2 as
E. pahinsintewakpa, but the metacone and metastylar
regions are not as reduced in these genera.
Macdonald (1963) originally described E.
pahinsintewakpa, the most plesiomorphic species of
Enhydrocyon, as Sunkahetanka pahinsintewaka, but X.
Wang (1994) placed this species into Enhydrocyon based

on the more derived cranial features compared to
Sunkahetanka geringensis. Since the Brooksville 2 spe-
cies has the above mentioned primitive features along
with a small third root above the metacone, it is appro-
priate to refer this material to E. pahinsintewakpa. How-
ever, with only a single tooth upon which to base this
assignment, it is appropriate to place a cf. (confero) be-
fore the trivial name.

Genus OSBORNODON Wang, 1994
Osbornodon wangi new species

ETYMOLOGY.- species, named for Xiaoming
HOLOTYPE.-UF 163693, associated RP3-
OCCURRENCE.-Brooksville 2 LF, Hernando
County, Florida; "medial" Arikareean UBI.
REFERRED MATERIAL.-UF 163696 partial
P4, UF163697 incisor, UF163724 m3, UF163743 Ca-
nine, UF175476 RP4, UF175477 LP3, UF175480
DIAGNOSIS.-Osbornodon wangi is larger than
the Whitneyan species of 0. sesnoni and 0. renjiei, and
falls within the small end of the size range of the
Hemingfordian 0. iamonensis (see Table 13). In gen-
eral, the dentition of O. wangi is not as robust as that of
0. iamonensis. 0. wangi can be separated from O.
iamonensis on the basis of the following plesiomorphic
features that are also found in the older species of
Osbornodon: the MI parastyle is more distinct; the
premolars are more slender in comparison to the molars
and the roots are more gracile.
DESCRIPTION.-The P3 is slender with a low
principle cusp, a posterior accessory cusp, and a cingular

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hemando County, Florida

cusp. The small P4 protocone is anteriorly placed and is
connected to the paracone by a slight crista that forms a
minute cuspule on the lingual slope of the paracone. A
thick ridge runs from the apex of the paracone to join
with the anterior cingulum at the anteromedial margin.
A lingual cingulum begins at the posterior of the meta-
cone and ends just anterior of the cusp of the paracone.
A slight buccal cingulum appears to start at the poste-
rior of the metacone and connect with the anterior cingu-
lum although part of the buccal margin is broken below
the carnassial notch, so the full extent of the cingulum is
based on other more complete Osbornodon P4s. The lin-
gual surface of the P4 is vertical and flat except for the
anterior portion.
The Ml is subquadrate in outline. The paracone is
taller than the metacone. Both cusps are subconical with
an anteroposterior ridge along the crests. The parastyle
almost equals the height of the metacone. It is low and
elongate but still distinct as a cusp. The protocone is
small and forms a posterobuccally directed ridge con-
necting with the metaconule by way of the post-
protocrista. The preprotocrista extends buccally to just
above the paracone. The metaconule continues pos-
teriorly to connect with the posterior cingulum. A well-
developed posteriorly directed internal cingulum is
present that is continuous with the slight anterior cingu-
lum. The hypocone is separated from the lingual cingu-
lum crest by a small transverse valley on the buccal slope
and a conical thickening of the cingulum. Posterior to
the protocone, the lingual cingulum reaches the same
height as the protocone until past the hypocone. After
the hypocone the lingual cingulum rises to the metacone
and joins the posterior cingulum.
The M2 is relatively large. The paracone is the larg-
est principle cusp but is less distinct from the metacone
than on the Ml. The metastylar region is absent and the
metacone is reduced compared to the M1. The hypocone
and lingual cingulum are large in relation to the size of
the M2.
DISCUSSION.-Osbornodon has the most exten-

Figure 14. Osbornodon wangi n. sp., UF163693, RP3-M2,

sive time distribution of the Hesperocyoninae. It ranges
from the Orellan to early Barstovian, although there is a
large time gap during the Arikareean that is devoid of
representation (X. Wang, 1994). In geographic range
Osbornodon is found from the Great Plains to southern
regions such as New Mexico and Florida. Osbornodon
is separated from other Hesperocyonines by two synapo-
morphies: "antero-posteriorly elongated (thus more quad-
rate) upper molars and enlargement of M2 and m2" (X.
Wang and Tedford, 1996:441). Most of the other Hespero-
cyonines tend to have reduced M2s like Enhydrocyon.
The common carnivore taxon in the famous Tho-
mas Farm locality ofGilchrist Co. Florida is Osbornodon
iamonensis. This sample was transferred to Osbornodon
from Cynodesmus by X. Wang (1994). In his review
Cynodesmus was restricted to two species existing from
the Whitneyan to the Early Arikareean. The Brooksville
2 material is broadly comparable to the Thomas Farm
species although there are some differences. Osborodon
iamonensis appears more robust and slightly more de-
rived than 0. wangi. Osborodon renjiei and 0. sesnoni
have a distinct parastyle while younger species lose the
distinct parastyle on the MI. The Brooksville 2
Osbornodon is intermediate in this feature. The P4 is
enlarged relative to the P3 in younger species but this is
not seen in the pre-Arikareean species. The Brooksville
2 Osbornodon has a P4 to P3 length ratio similar to O.
iamonensis, but the P3 has narrower proportions, as in
the older species.
Osbornodon is a conservative genus in dental mor-
phology. X. Wang and Tedford (1996) use small size
difference and more slender premolars to distinguish O.
sesnoni from 0. renjiei. In younger species of
Osbornodon cranial features become more important in
diagnosing species. In light of the differences outlined
above, and following the taxonomy of X. Wang (1994),
the naming a new species is warranted.
Most of the Arikareean has no representative of the
Osbornodon lineage. Therefore, the Brooksville 2 sample
represents a significant new species that helps fill the


TABLE 12.-Dental measurements of Osbornodon wangi n. sp. with closest species of Osbornodon
for comparison, from X. Wang 1994. Measurements made following X. Wang 1994:17, Fig.6.

Osbornodon sesnoni Osbornodon wangi Osbornodon iamonensis
P4 ap 13.35 13.0-13.7 15.0 18.0 14.6-19.6
tra 8.6 x 8.0 10.4 8.5-12.2
trp 15.4 x 16.4 19.44 15.7-21.8

M1 ap 10.1 x 10.6 11.94 10.0-13.3
tra 13.6 x 15.2 16.93 13.7-18.7
trp 11.6 x 12.2 14.43 12.2-16.2

M2 ap 6.05 5.8-6.3 6.3 7.0 5.6-8.0
tra 9.85 9.4-10.3 10.4 11.0 8.7-13.0
trp 8.3 8.0-8.6 8.8 9.66 7.9-12.0

gap between 0. sesnoni and younger, larger species of

Subfamily BOROPHAGINAE Simpson, 1948
Tribe PHLAOCYONINI X. Wang et al. 1999
Genus PHLAOCYON Matthew, 1899
Phlaocyon taylori new species
Figures 15, 16, Table 13

ETYMOLOGY.- species, named for Eric Taylor
HOLOTYPE.-UF163524 Lml
OCCURRENCE.-Brooksville 2 LF, Hemando
Co., Cow House Slough LF, Hillsborough Co., Florida;
"medial" Arikareean UBI
UF163502 LM2, UF173503 Rml, UF163504 Lm2,
UF163505 RMI, UF163506 RM1, UF163507 RMI,
UF163508 Rml, UF163509 RM1, UF163510 LMI,
UF163511 Lm2, UF163512 Lml, UF163513 RMI,
UF163514 LMI, UF163515 Rm2, UF163516 Rml,
UF163517 RP4, UF163519 LP4, UF163520 LM1,
UF163521 RMI, UF163522 RM2, UF163525 Lml
trigonid, UF163526 Lml, UF163527 RM2, UF163528
ml talonid, UF163529 Rp3, UF163531 Rp4, UF163533
LP4, UF163534 RP4, UF163701 RM2, UF163716 m3,
UF163720P4,UF163721 P4,UF163724m3,UF163741
canine, UF163818 LM1, UF163819 LM1; from
Brooksville 2 LF. UF40149 Lp4, UF40151 canine; from
Cow House Slough LF.
DIAGNOSIS.-Smallest species of the genus
(Fig. 15). Closest in size to P achoros. Well-developed

metaconule and paraconule on MI. Metaconule twinned
on M like P achoros and P multicuspus. Paraconule
variably twinned. Distinguishable from P achoros by
the following features: absent to weakly developed hypo-
cone on P4, lingual cingulum on Ml cuspidate, protostylid
absent to minute on ml, accessory cusps on posterior
cristid and cristid obliquid of talonid ml, and mesoconid
present. Differs from P multicuspus by its much smaller
size and absence of conical hypocone on M2.
DESCRIPTION.-On P4 the paracone forms a
tall enlarged cone. The bladelike metacone is half the
height of the paracone with a small division of the cen-
tral crest. The lingual cingulum is well-developed. The
hypocone varies from a thickening in the lingual cingu-
lum (Fig. 16:A) to a minute, but distinct, cusp (Fig. 16:C,
D). The anteriorly placed protocone forms a distinct high
cusp. A small ridge runs from the apex of the paracone
to a low elongate central parastyle that is connected to
an anterior cingulum. A slight buccal cingulum is present
that ends at the carnassial notch.
The upper molars are cuspidate and quadrate. The
paracone is the tallest cusp, followed by the metacone
and then the protocone. The paracone and metacone are
in line, elongated transversely, and connected together
by a "V"-shaped centrocrista. A thick buccal cingulum
is present that becomes wider in the parastylar region.
On M a twinned metaconule is always present and is
tripled in some specimens (Fig. 16:G). The paraconule is
well developed and usually doubled (Fig.16:E, F, G).
The post-hypocrista extends from the metacone to the
posteriorly directed hypocone. The lingual cingulum is
prominent with small cuspules on the crest. It joins the

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

anterior cingulum that is continuous with the buccal cin-
gulum at the anterobuccal margin. M2 has a reduced
metacone and the metastylar region is absent which makes
the buccal margin slant posterolingually. The metaconule
is small relative to the MI and not doubled. The
paraconule is also reduced. The lingual cingulum is com-
pressed buccally lingual to the protocone, giving M2 a
more oval shape than MI (Fig. 16:A). A distinct hypo-
cone is not present.
Lower p4 is slender and tall (Fig. 16:H, I). A small
cingular cusp lies on the anterior and another on the pos-
terior margin of the tooth. A weak accessory cusp sits
high on the posterior slope of the principle cusp. A slight
buccal cingulid is present.
Lower ml is distinctive due to its hypocarnivorous
adaptations (Fig. 16:J, K, L). The protoconid is the domi-
nant trigonid cusp. The paraconid lies almost directly
anterior of the metaconid. The buccal side of the para-
cone forms a blade that meets the protoconid at the car-
nassial notch. The lingual portion of the paraconid is
low and open. A slight buccal cingulum runs along the
trigonid, beginning just below the paraconid and curv-
ing upward toward the posterior protoconid. A weak
protostylid occupies the buccal slope of the protocone
on the holotype specimen, but is not found on the other
mis. The talonid is basined and cuspidate. A distinct
mesoconid lies on the posterior slope of the metaconid.

The entoconid is the dominant cusp and also the most
conical of the talonid cusps. Two to three accessory cusps
occur on the posterior cristid and the crista obliqua. The
hypocone forms the raised posterior buccal corner of the
crista obliqua. The crista obliqua is connected to the trigo-
nid wall just buccal to the protoconid-metaconid valley.
A deciduous p4 was recovered and is of proper size
and morphology to be assignable to Phlaocyon taylori.
This dp4 is gracile compared to the permanent ml. The
cristid obliquid is absent, so that the hypocone lies next
to the trigonid wall. The posterior cristid is slanted
posterobuccally with four accessory cusps that enlarge
successively toward the posterior end of the premolar.
The m2 differs from ml in its shorter anteroposter-
ior ratio, enlarged buccal cingulid, small protostylid, and
paraconid diminished to the size of the buccal cingulid.
All the other cusps are similar to the ml but changed in
proportional size (Fig. 16:M). The m3 is small and cir-
cular in outline. Only the metacone and protocone are
discernible. They form an indistinct transverse ridge.
DISCUSSION.-In morphology and size
Phlaocyon taylori n.sp. is close to what Frailey (1979)
described as "Bassariscops achoros from the Buda LF.
In their review of the Borophaginae, X. Wang et al. (1999)
recognize two genera within Frailey's hypodigm of B.
achoros. To the genus Phlaocyon they assigned the hypo-
carnivorous specimens, the holotype of B. achoros, and

i Ph- fonlaylsri
- PiMaocyon achs
...... .. phlaocyonwi li i
.. i.. P Nmaocyon multacu us f \
-. .o A 'm/
.. ^/ :/ '. \
-.,-- 5
.. ,L \%
Ss '"
'5 ,


M2 width

M2 length

MI length P4 width P4 length

MI width

Figure 15. Log differentials of dental measurements among species of Phlaocyon compared to Archaeocyon pavidus as
standard. (Modified from X. Wang et al. 1999.)





i N




Figure 16. Phlaocyon taylori n. sp., x4.5, scale bar =1 mm. (A) composite, UF163522, RM2; UF163521, RMI; UF163534;
occlusal view; (B) composite, buccal view; (C) UF163517, partial RP4, lingual view; (D) UF163517, occlusal view; (E)
UF163513, RM1; (F) UF163520, LMI; (G) UF163514, RMI; (H) UF163720, Rp4, occlusal view; (I) UF163720, buccal view;
(J) UF163524, Lml, holotype, occlusal view; (K) UF163524, buccal view; (L) UF163524, lingual view; (M) UF163511, Rm2;
(N) UF163700, Rdp4, occlusal view; (0) UF163700, occlusal view.

an ml Frailey had referred to Cynarctoides. These assign-
ments are based on their cladistic analysis of the genus,
specifically on shared synapomorphies of these speci-
mens with P multicuspus. They placed the more meso-
carnivorous teeth from the Buda LF in Cynartoides lemur.
Peterson (1924) originally included "Bassariscops"
in Phlaocyon when it was considered a procyonid be-
cause of its hypocarnivorous dentition. Hough (1948)
later demonstrated that Phlaocyon ear region was es-
sentially canid and Frailey (1979) further solidified this

when he compared the anatomy of the petrosal bone of
"Bassariscops" achoros with those of other taxa and
determined that its relationship was clearly with the
canids. X. Wang et al. (1999) place Phlaocyon achoros
within the borophagine canids as part of a
hypocarnivorous clade encompassing Cynarctoides
McGrew, 1938; "Bassariscops" Peterson, 1924; and
"Aletocyon" Romer and Sutton, 1927.
In the new species diagnosis X. Wang et al. (1999)
suggest that P achoros is allied to the Hemingfordian P





HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

multicuspus. Even though P multicuspus is much larger,
the taxa share several derived features. The twinned
metaconule is also shared with Phlaocyon taylori. These
three species are the only ones to possess this character
in the Phlaocyon clade, suggesting a separate, more
hypocarnivorous lineage within the clade. On the other
hand, proposing a direct lineage between these taxa is
problematic due to the fact that each possesses derived
unique characters of its own and the similarities may be
convergent. As stated in the species diagnosis the
Brooksville 2 Phlaocyon has even more cuspidate mo-
lars than the younger P achoros, whereas P achoros
and P multicuspus each has a much larger hypocone on
P4 and a well defined protostylid on ml. This was not
observed in Phlaocyon taylori n. sp.
Phlaocyon is a common omnivorous member of the
Arikareean faunas of Florida, occurring in four of the
seven known localities. Along with Buda and
Brooksville, SB-IA has a P4 (TRO 392) that Frailey
(1978) described as Phlaocyon sp. This upper carnas-
sial has a well-developed hypocone, and therefore prob-

ably represents another species ofPhlaocyon, rather than
the Brooksville 2 LF species. In addition, the sparse car-
nivore material from Cow House Slough LF indicates
the presence of Phlaocyon taylori n. sp. on the basis of
UF40149, a left p4, and possibly UF40151, a canine.
Specimens tentatively labeled Phlaocyon (UF163892-
96) in the Thomas Farm collection possibly belong to
this genus, but more study is needed. If the Thomas Farm
material is Phlaocyon, it may represent a species de-
rived from a lineage that begins with Phlaocyon taylori.
This lineage follows a trend of gaining complexity in
the P4 and losing complexity in the Ml and M2. The
trend might be used in the future for biochronological
refinement of stages in the Arikareean and


Correlation within the Arikareean NALMA presents
some difficulty because there is no currently accepted
biochronology for the "age" that spans the interval from

TABLE 13.-Dental measurements of Phlaocyon taylori n. sp. from Brooksville 2 LF and P achoros
from the Buda LF for comparison.

Phlaocyon tavlori
P4 (UF163533) (UF163534)
ap 6.2 6.0
tr 3.8 3.3

ap 11 5.14 4.6-5.8 0.36
tr 10 6.19 5.7-6.9 0.38








Phlaocyon achoros
(UF18389) (UF22778)
6.8 4.1
7.1 4.4













30-19 Ma. The Arikareean was initially characterized
by an assemblage largely collected from the upper
Arikaree Group, but including some mammalian genera
typical of the lower Arikareean. Tedford et al. (1987)
defined the late Arikareean (c.22-18.8 Ma; new dates of
MacFadden and Hunt, 1998) by the earliest appearance
of Oligobunis, Zodiolestes, Daphoenodon, and
Menoceras. They also used data collected by research-
ers on the basal Arikaree formations to define the begin-
ning of the Arikareean. The sparseness of well-dated
faunas in the intervening time period did not provide
adequate data to establish a detailed biochronology for
the greater part of Arikareean time. Woodburne and
Swisher (1995) later proposed a division of the
Arikareean into four "zones" based on mammal dispersal
events: Arl-Ar2 (30-27.5 Ma), Ar2-Ar3 (27.5-24 Ma),
Ar3-Ar4 (24-21 Ma), and Ar4-Hemingfordian (21-
19Ma). However, they did not define and characterize
these new "zones" and they are not tied to rock units, so
biostratigraphic definition is lacking. Later, study of the
Whitneyan-Arikareean transition by Tedford et al. (1996)
led to a redefined biochronology of the early Arikareean.
They proposed that the early Arikareean be defined by
the first appearance of Plesiosminthus and character-
ized by the appearances of several autochthonous mam-
mals including Nanotragulus loomisi, Palaeolagus
philoi, and Palaeocastor nebraskensis. Furthermore,
they suggested that the faunal turnover event at ca. 28
Ma, characterized by the first appearance ofParvericius,
the eomyid Pseudotheridomys, Promartes, and the
chalicothereMoropus, could be used to define the begin-
ning of the medial Arikareean "age." The medial inter-
val, so defined would cover the time from 28 to 22 Ma.
To call this "medial Arikareean" would imply an ap-
proximate time equivalence to the early and late Arika-
reean intervals. What is present is a long interval be-
tween faunal biochrons at the base of the Arikaree Group
and those in the upper part. This long interval is biochron-
ologically undefined and I therefore use "medial"
Arikareean Undefined Biochronologic Interval (UBI) for
this time period. For purposes of discussion, where other
authors have used either the Woodburne and Swisher
(1995) chronology or the Tedford et al. (1987) chronology,
I have retained the original terms used by those authors.
The most recent discussion of the relationship of
the Florida and Gulf Coast Arikareean age sites was by
Albright (1998). Based on his faunal analysis, Albright
placed Cow House Slough and White Springs in Ar2
(25 and 24.4 Ma, respectively) as the oldest of Florida
faunas. Buda, Franklin Phosphate Pit No. 2, and Brooks-
ville 1 were positioned together as slightly younger (Ar2-
Ar3: 24-23.5 Ma). SB-lA was correlated with the begin-


ning of Ar3 (22 Ma). The Martin-Anthony oreodont was
separated from the fauna (Martin-Anthony LF) found in
the clastics above the oreodont and placed as slightly
younger than SB-IA at the end of Ar3 (21 Ma). The
Martin-Anthony LF itself was placed in Ar4 (19-20 Ma).
Before Albright's correlation of the Florida
Arikareean, Morgan (1993) attempted to address the
problem of early mammalian biochronology in Florida
by studying those sites that placed terrestrial mammals
within marine sedimentary contexts, expanding on the
work of Tedford and Hunter (1984). He considered Cow
House Slough LF and, particularly, the White Springs
LF, since the latter was found in nearshore marine units.
Within the context of Tedford et al.'s (1987) review of
the Arikareean biochron, Morgan correlated both fau-
nas to the early late Arikareean. Morgan (1993) also
placed the previously described (MacFadden, 1980)
Martin-Anthony LF in the late late Arikareean. This as-
signment of White Springs LF to the early late Arikareean
represented a change from Morgan's (1989) earlier cor-
relation of the marine sediments at White Springs to the
upper N4 and lower N5 subtropical foraminiferal zones
(22-21 Ma). New evidence in the form of 87Sr/86Sr
dating of mollusk shells from the White Springs locality
gave an "age" of 24.4 Ma (Jones et al., 1993).
The correlation and age determination of most of
Florida's early vertebrate sites present several additional
problems. Most of Florida's late Oligocene-early Mi-
ocene terrestrial vertebrate fossils are found in karst de-
posits and thus represent a unique style of sedimentation
unlike the Arikareean volcaniclastics that blanket the
Great Plains and John Day regions. Lithologic correla-
tion, magnetostratigraphy, and radioisotopic dating fre-
quently can be applied to these western faunas but are
often of little use in correlation of the earliest Florida
sites. Karst preservation may also introduce another prob-
lem in the form of distinct paleoecological biases. Karst
depositions may take place on a small scale in any one
of a variety of habitats from pine/oak forests to ham-
mocks, swamps, and lakes, and can therefore produce a
fauna that is environmentally specific, rather than a broad
sampling of taxa. Faunas that appear to be temporally
disjunct because of differing taxa may actually be of simi-
lar age but come from distinct ecotones. Karst forma-
tion creates large open sinkholes, lakes, caves, and solu-
tion pipes, which can result in a preserved thana-
toceonosis representing only a fraction of the true eco-
logical diversity. Karst features may filter out those ani-
mals that are too large to enter the caves, or use of these
solution features may be dominated by a particular group
of animals that may exclude utilization by other animals.
In view of the above problems, and even consider-

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hemando County, Florida

ing the small degree of endemism of the Florida
Arikareean taxa identified by Albright (1998), determi-
nation of the age of Florida's early "sinkhole" faunas is
best done by comparison with the well dated western
faunas. This comparison, as well as comparison with
Florida localities such as the Hemingfordian Thomas
Farm LF, has led to a consensus of Florida sites repre-
senting life during the Arikareean NALMA (Frailey,
1978, 1979; MacFadden and Webb, 1982; Tedford et
al., 1987; Morgan, 1993; Albright, 1998). However, de-
termination of the relative positions of the Arikareean
Florida faunas, and their placement within the subdivi-
sions of the Arikareean has not been universally accepted
or adequately constrained.
The Brooksville 2 LF contains some taxa that di-
rectly compare with western faunas of the Arikareean.
The Brooksville 2 LF includes the hedgehog Parvericius
whose first appearance is used to define the beginning of
Tedford et al.'s (1996) "medial" Arikareean. Entop-
tychine rodents, represented in the Brooksville 2 LF by
an indeterminate genus, also do not appear until the be-
ginning of the "medial" Arikareean (Tedford et al., 1996).
Several taxa present in the fauna serve to constrain
the youngest possible date for the fauna. Centetodon has
its latest occurrence in the McCann Canyon LF of the
Great Plains (Korth, 1992), which was placed in the late
Arikareean of Tedford et al. (1987). Agnotocastor is last
recorded in North America at around 27.5 Ma (Xu,
1996). Nanotragulus has its last appearance in the late
late Arikareean (Ar4). However, Frailey (1979) identi-
fied an earlier form, Nanotragulus loomisi, as a key el-
ement in the Buda LF and compared it directly to the
same species in the Arikareean of the Great Plains.
Nanotragulus loomisi is the holotype species of
Nanotragulus; a sample of this species from the Arikaree
Group of the Muddy Creek region, Hartville Uplift, south-
eastern Wyoming (including the holotype), averages sig-
nificantly smaller than individuals ofNanotragulus from
the upper Arikaree Group of northwest Nebraska (R.
M. Hunt, pers. comm., February 1999). Because the
species of the genus appear to increase in size through
time, N. loomisi may be a potentially useful correlator
within the Arikareean NALMA. Its presence in the
Brooksville 2 LF suggests a pre-late Arikareean age.
Other taxa that compare on the generic level to the
Great Plains and also indicate a younger than early
Arikareean age for the fauna are Megalagus abaconis,
Miohippus sp., Herpetotherium sp., Enhydrocyon cf.
pahinsintewakpa, and Osbornodon wangi. Prothero and
Emry (1996) listed Megalagus as having its last occur-
rence at the end of the Whitneyan. This range limitation
is in error as two Arikareean species have been described,

M. primitivus from the Harrison Formation by Dawson
(1967) and M dawsoni from the Fort Logan Formation
of Montana by Black (1961). Megalagus abaconis is
derived in comparison with Megalagus from the early
Arikareean which have larger buccal roots, less molari-
form premolars, and less hypsodont cheek teeth.
Miohippus sp. from Brooksville 2 LF and Cow House
Slough is advanced for the genus and is probably a tran-
sitional form between Miohippus of the early Arikareean
and later Archaeohippus. Although Herpetotherium is
different in dental morphology from the Arikareean H.
young of the Great Plains, it is of similar size and fol-
lows the general trend for Herpetotherium in size reduc-
tion from oldest to youngest species. Also, comparison
of this Herpetotherium with Herpetotherium from the
Whitneyan or early Arikareean 1-75 LF indicates an an-
cestor-descendant relationship. The taxonomy and rela-
tionships of Florida marsupials will be discussed in Hayes
and Wolff, in progress.
The presence of Osbornodon in the Brooksville 2
LF is one of the few genera of the Florida Arikareean
that ranges into the Hemingfordian. A morphological
comparison between the Brooksville 2 LF Osbornodon
wangi and the Thomas Farm Osbornodon iamonensis
shows that the Brooksville 2 LF sample is more
plesiomorphic than the Thomas Farm species and thus
probably older. In other characters Osbornodon wangi
is more derived and larger than either 0. sesnoni or O.
renjiei, which have their last appearance at the end of
the Whitneyan (X. Wang and Tedford, 1996).
Based on the foregoing the new Brooksville 2 LF
can be placed in the "medial" Arikareean UBI. The ques-
tion remains as to where in this very long interval the
Brooksville 2 LF fits and where the other Florida
Arikareean sites belong in relation to it.
Table 14 is a comparison of the Florida Arikareean
faunas and the Arikareean-aged Toledo Bend LF from
Texas. Note that all of these faunas have genera and spe-
cies that are not shared with any of the others, as well as
many taxa that are shared by only a few of the other
faunas. The importance of the Brooksville 2 LF is that
this fauna holds a greater diversity of taxa that link all of
the other faunas, thus strengthening their correlation with
one another.
Of the Florida Arikareean sites, Brooksville 2 LF
has the greatest homotaxial similarity with Cow House
Slough and SB-IA. However, only four taxa are shared
between all three faunas: Proheteromys, Phlaocyon,
Entoptychinae, and Nothokemas. Cow House Slough is
the most similar to the Brooksville 2 LF. It has a total of
eight taxa that are shared with Brooksville 2 LF. Beside
the above, they share Parvericius, Agnotocastor,


TABLE 14.-Mammalian taxa from Arikareean Florida Sites and the Toledo Bend LF.

Brooks- Cow White Franklin Martin- Toledo
1-75 ville 2 SB-1A House S. Buda Spring Phos. Pit Anthony Bend

Palaeogale minutus
Phlaocyon taylori
Phlaocyon achoros
Phlaocyon sp.
Anchippus texanus
Nanotragulus loomisi
Nanotragulus sp.


x x

x x

x x

x x

x x


X? x

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

Megalagus, andMiohippus. Each of these genera is rep-
resented by the same species in both faunas. SB-i A shares
seven genera with Brooksville including Herpetotherium
sp., Palaeogale minutus, and Arikarictis chapini that
are not found in Cow House Slough. The Buda LF con-
tains six of the same genera as Brooksville 2 and two to
three of the same species. White Springs LF has four to
five of the same genera but none of the same species.
Franklin Phosphate Pit No. 2 and the Toledo Bend Local
Faunas share Proheteromys with Brooksville 2 LF, but
only Franklin Phosphate Pit No. 2 LF has Nothokemas
and only Toledo Bend LF has Miohippus.
The Martin-Anthony LF has the least in common
with Brooksville 2 or any of the other faunas. I should
also note here that the Brooksville 1 LF is not included
in this discussion or Table 14 because it consists of only
4 teeth that do not have adequate locality data; Patton
(1967) only reported that they originated from somewhere
in the Brooksville area.
While it is clear that the Brooksville 2, Cow House
Slough, Buda, SB-lA, Franklin Phosphate Pit, and To-
ledo Bend Local Faunas belong to an interval within the
Arikareean age from 28 to 22 Ma, it is more difficult to
place these Arikareean faunas in relation to one another
considering that the faunal discrepancies between the sites
may be merely ecological.
Two taxa are found in almost all Florida Arikareean
faunas, Nothokemas and Proheteromys. Recognition of
Nothokemas in many of the faunas indicates that this
genus is one of the few reliable Arikareean age indica-
tors in Florida. However, at present, it does not separate
the Arikareean faunas from each other. (Further study
of these samples may indicate evolutionary differences
that are reliable for use in correlation.) Proheteromys is
shared by all Arikareean sites as well as the Thomas
Farm locality and analysis of these rodents may deter-
mine if there are species level differences that can be
used as age indicators. Such a study of Proheteromys
lay beyond the scope of this study.
The Phlaocyon species in these faunas are some-
what enigmatic. Dental characters distinguish Phlaocyon
achoros at Buda from P taylori in the Brooksville 2 LF.
The Brooksville 2 species has more cuspidate molars,
but lacks the developed hypocone on the P4 and the de-
veloped protostylid on ml of P achoros. The Thomas
Farm Phlaocyon has less complex molar teeth but more
complex carnassials, suggesting an ancestor-descendant
relationship may exist with the Brooksville 2 species that
could help determine age. Phlaocyon sp. from SB-IA
has a developed hypocone on P4 like P achoros, and yet

the SB-lA LF as a whole is much more comparable to
Brooksville 2 than to Buda.
The beaver Agnotocastor from Brooksville 2, in
comparison with the Toledo Bend castorid, Neatocastor,
is more pleisomorphic. The Toledo Bend and White
Springs beaver is taller crowned and the fossettids are
more complex. The White Springs beaver also appears
to belong to Neatocastor, although the material is not
diagnostic. In the Great Plains Agnotocastor disappears
by 27.5 Ma (Xu, 1996). Korth and Emry (1997) pro-
pose an ancestor-descendant relationship between
Agnotocastor and Neatocastor that would support an
earlier age assignment for Brooksville 2 in relation to
Toledo Bend or White Springs.
Entoptychine rodents are found at Cow House
Slough and SB-IA as well as Brooksville 2. Rodents are
common in the rest of the faunas. The size bias that may
reduce the size of larger taxa does not seem to hinder
sampling of rodent taxa. Thus the presence of entopty-
chines may indicate an earlier age for these three faunas.
Rabbits are found in only two other Florida Ari-
kareean faunas, Cow House Slough and White Springs,
so their use in correlation is limited. Megalagus abaconis
is the same taxon as the Cow House Slough lagomorph,
but the White Springs rabbit is smaller and more similar
to Archaeolagus.
In summary, species to species comparison, while
not definitive, seems to place Brooksville 2 LF as tem-
porally older than Buda, Franklin Phosphate Pit, White
Springs, and Toledo Bend. Brooksville 2 LF, Cow House
Slough, and SB-IA are probably close in age, and may
represent the "early-medial" Arikareean UBI (28-26 Ma)
based on comparisons with Great Plains faunas. This
age assignment for the Brooksville 2 LF is also supported
by sea level fall during the earlier late Oligocene
(Woodbume and Swisher, 1995, Fig.3). Karst forma-
tion and exposure of the Florida Platform would have
been greatest at this time (28-25 Ma). Buda, Franklin
Phosphate Pit No. 2, and Toledo Bend represent a younger
time period or the "late-medial" Arikareean UBI. The
White Springs LF was numerically dated using Sr87/
Sr86 at 24.4 Ma (Jones et al., 1993) and is slightly older
than Buda and Toledo Bend. My correlations of the Gulf
Coast faunas of Arikareean age agree with that of
Albright (1998), with two exceptions. Based on the simi-
larity with the Brooksville 2 LF, SB-IA is moved into an
older position, earlier than White Springs (26-25 Ma).
Also, Cow House Slough is considered temporally close
to the Brooksville 2 LF and therefore correlated to the
"early-medial" Arikareean UBI (28-26 Ma).


The Brooksville 2 Local Fauna contributes a sub-
stantial new sample to the growing knowledge of
Arikareean mammal faunas in the Gulf Coastal region.
It is the most diverse Arikareean fauna now known from
Florida, rivaling the Toledo Bend Local Fauna from
Texas. The diverse taxa in the fauna aid in correlating
many of the Arikareean faunas in Florida and provides
for a more precise characterization of the Arikareean
NALMA in Florida.
The Brooksville 2 LF formed during the late Oli-
gocene sea level low and at about the time when the Ocala
Platform began to arch upward. This led to karst ero-
sion of the early Oligocene Suwannee Limestone. Some
terrestrial vertebrates, such as bats and small carnivores,
lived and died in these caves and sinkholes. As the karst
terrain evolved, other vertebrate remains were transported
and redeposited subaqueously in the fissure system form-









HW 0 u
D -


...................Ar3 22




ing the five clay pockets (sites IA-1E) from which the
Brooksville 2 LF was recovered.
Faunal correlations place the age of the Brooksville
2 LF within the earlier part of the "medial" Arikareean
NALMA UBI (28-26 ma). In Florida, The Brooksville 2
LF shares the most genera and species with the
undescribed Cow House Slough LF (eight) followed by
the SB-IA LF (seven). These faunas are considered cor-
relative. The Buda LF, Franklin Phosphate Pit No. 2 LF,
and Toledo Bend LF are placed together in a later part of
the "medial" Arikareean.
Taxa from the Brooksville 2 LF representing first
occurrences in Florida include Agnotocastor,Megalagus,
Enhydrocyon, and an entoptychine rodent. New species
are described for Megalagus, Osbornodon, and Phla-
ocyon. Two new genera and species of musteloids are
described, Arikarictis chapini and Acheronictis webbi.
The marsupial in the fauna, Herpetotherium, may
be a new lineage of this genus in the Gulf Coastal Plain.
The 1-75 Herpetotherium is an intermediate species be-





- 4 +

18 Thomas FarmLF

White Springs LF

? SB-1A LF ?


Ar 26

Martin-Anthony LF

Franklin Buda

Cow House S. LF


Toledo Bend

Figure 17. Correlation chart showing age and stratigraphic position of Arikareean and Hemingfordian local faunas of Florida
and the Texas Toledo Bend LF. Mammal "ages" and boundaries follows Woodburne and Swisher (1995), Tedford et al. (1996)
and MacFadden and Hunt (1998).

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hemando County, Florida

tween H. valens and Herpetotherium n. sp. from Brooks-
ville 2 LF.
Two insectivore taxa, Centetodon magnus and the
hedgehog Parvericius, are well represented in the
Brookville 2 LF for the first time in the Gulf Coastal
Region. The described Amphechinus sp. from Buda is
here placed in Parvericius. Parvericius is one of three
taxa in the Brooksville 2 LF directly comparable to Great
Plains faunas.
Agnotocastor sp. extends the geographic range of
this genus into Florida, it is known from the Great Plains
and last occurs there in the early Arikareean (27.5MA).
The leporid, Megalagus abaconis, is an advanced form
of this genus, similar to the Arikareean M. dawsoni from
Montana. Both Agnotocastor and Megalagus are found
in the Cow House Slough LF The White Spring LF
castorid and lagomorph are more derived species, per-
haps Neatocastor and Archaeolagus. After White
Springs LF, rabbits and beavers are curiously absent from
the Florida record until the Barstovian.
Entoptychine rodents are identified for the first time
in Florida. They occur in the Brooksville 2 LF, Cow
House Slough, and SB-IA LF. This species is allied to
Gregorymys and Entoptychus from the Great Plains and
John Day faunas but is a distinct Gulf Coast species if
not genus.
Albright (1996) described Palaeogale from the
Arikareean of the Gulf Coast on the basis of a single
upper carnassial. In light of more conclusive material
from Brooksville 2 that can be assigned with confidence
to Palaeogale, the Toledo Bend specimen is referred to
the Mustelidae. The Palaeogale material is assigned to
P minute and extends downward the known size-range
of this genus.
The two musteloids in the fauna indicate that
musteloid diversity within the Arikareean of North
America is greater than formerly recognized. They help

fill the gap between the Chadronian Mustelavus and the
appearance of Promartes and Oligobunis in the
Arikareean. Acheronictis webbi is the smallest muste-
loid in the fauna and may be similar to "?Plesictis" sp.
described by MacDonald (1970) from the Wounded Knee
Fauna. Arikarictis chapini is similar in morphology to
early European genera of Musteloids.
Phlaocyon is recognized as a common raccoon-sized
canid in the Arikareean of Florida. X. Wang et al. (1999)
refer Frailey's (1979) "Bassariscops" achoros from Buda
LF to Phlaocyon. The sample from Brooksville 2 LF is
a new species, P taylori, separate from and smaller than
P achoros.
Osbornodon wangi fills part of the time gap be-
tween the Whitneyan 0. sesnoni and the Hemingfordian
0. iamonensis. The presence of Enhydrocyon in the fauna
marks a first occurrence of this taxon since Tedford and
Frailey (1976) referred the Thomas Farm "Enhydrocyon"
to Euoplocyon.
Recognition of Nothokemas from almost all the
Florida Arikareean faunas makes this wider-ranging
taxon important for general correlation of the Florida
Arikareean. The smallest artiodactyl in the Brooksville
2 LF, Nanotragulus loomisi, is important as one of the
species directly comparable to the Great Plains.
The composition of the Brooksville 2 LF reinforces
the view that during the Arikareean the faunas of the
Gulf Coastal Plain exhibited a high degree of provin-
cialism. The newly described species represent smaller
forms of their respective genera, a product of Bergman's
rule or an indication of island dwarfing if Florida was
still separate from the continent. However, many of the
taxa have generic or subfamilial counterparts found in
the Great Plains, proving that what Simpson surmised
in 1930 is still correct: early Florida shows "Geographic
separation (but not isolation) from the Great Plains"



Albright, L. B. 1991. The vertebrate paleontology,
taphonomy, and paleoecology of a new early Miocene
site in the Fleming Formation near Toledo Bend Dam,
Newton County, Texas and its importance to Gulf Coast
biostratigraphy. M.S. thesis. Louisiana State Univ.,
Baton Rouge. 319 pp.
1994. Lower vertebrates from an Arikareean (ear-
liest Miocene) fauna near Toledo Bend Dam, Newton
County, Texas. J. Paleontol. 68:1131-1145.
1996. Insectivores, rodents, and carnivores of the
Toledo Bend local fauna: An Arikareean (earliest Mi-
ocene) assemblage from the Texas Coastal Plain. J.
Vert. Paleontol. 16:458-473.
1998. The Arikareean land mammal age in Texas
and Florida: Southern extension of Great Plains faunas
and Gulf Coastal Plain endemism. Pp.167-183 in Depo-
sitional Environments, Lithostratigraphy, and Bios-
tratigraphy of the White River and Arikaree Groups
(Late Eocene to Early Miocene, North America), ed.
D. O. Terry, H. E. LaGarry, and R. M. Hunt, Jr. Geol.
Soc. Amer. Spec. Pap. 325.
Baskin, J. A. 1998a. Procyonidae. Chapter 8, pp. 144-151
in Evolution of Tertiary Mammals of North America,
Volume 1: Terrestrial Carnivores, Ungulates, and Un-
gulate Like Mammals, ed. C. M. Janus, K. M. Scott,
and L. L. Jacobs. New York: Cambridge University
1998b. Mustelidae. Chapter 9, pp. 152-173 in Evo-
lution of Tertiary Mammals of North America, Volume
1: Terrestrial Carnivores, Ungulates, and Ungulate Like
Mammals, ed. C. M. Janus, K. M. Scott, and L. L. Jacobs.
New York: Cambridge University Press.
and R. H. Tedford. 1996. Small arctoid and feliform
carnivorans. Pp.486- 497 in The Terrestrial Eocene-
Oligocene Transition in North America, ed. D. R.
Prothero and R. J. Emry. New York: Cambridge
University Press.
Behrensmeyer, A. K., and R.W. Hook. 1992. Paleo-
environmental contexts and taphonomic modes.
Chapter 2, pp.15-136 in Terrestrial Ecosystems through
Time: Evolutionary Paleoecology of Terrestrial Plants
and Animals, ed. A. K. Behrensmeyer, J. D. Damuth,
W. A. DiMichele, R. Potts, H-D. Sues, and S. L. Wing.
Chicago: University of Chicago Press.
Black, C. C. 1961. Rodents and lagomorphs from the Mi-
ocene Fort Logan and Deep River Formations of Mon-
tana. Postilla, Yale Peabody Mus. 48:1-20.
Bonis, L. de. 1981. Contribution a i'6tude du genre Palaeo-
gale Meyer (Mammalia, Carnivora). Annales de
Paldontologie (Vertebrds) 67:37-56.

_ and Y. Guinot. 1987. Le gisement de Vertdbrds de
Thezels (Lot) et la limited Oligo-Mioc6ne dans les
formations continentales du bassin d'Aquitaine.
Miinchner Geowissenschaftliche Abhandlungen, Reihe
A 10: 49-57.
Clark, J. 1936. Mustelavuspriscus gen. et sp. nov. (Manu-
script) in The mammalian fauna of the White River
Oligocene, Part 1.Insectivora and Carnivora, ed. W. B.
Scott and G. L. Jepsen. Amer. Phil. Soc. Trans., n.s.
Cirot, E., and M. Wolsan. 1994. Late Oligocene Amphictids
(Mammalia: Carnivora) from la Millogue, Aquitane
Basin, France. Geobios 28:757-767.
Dawson, M. R. 1958. Later Tertiary Leporidae of North
America. Univ. Kansas Palaeontol. Contr., Vertebrata,
Article 6:1-75.
.1967. Lagomorph history and the stratigraphic
record. Essays in Paleontology and Stratigraphy,
Raymond C. Moore Commemorative Volume. Univ.
Kansas Dept. Geol. Spec. Publ. 2:287-316.
Dehm, R. 1950. Die Raubtiere aus dem Mittel-Miocan
(Burdigalium) von Wintershof-West bei Eichstatt in
Bayern. Abhandlungen derBayerischen Akademie der
Wissenschaften, mathemnatisch-naturwissenschaftliche
Klasse, Neue Folge 58:1-141.
Ewer, R. F. 1973. The Carnivores. Ithaca NY: Cornell
University Press. 494 pp.
Flynn, J.J., and H. Galiano. 1982. Phylogeny of early Ter-
tiary Carnivora, with a description of a new species of
Protictis from the middle Eocene of Northwestern
Wyoming. Amer. Mus. Nov. 2725:1-64.
_ N. A. Neff, and R. H. Tedford. 1988. Phylogeny of
the Carnivora. Pp.73-116 in The Phylogeny and
Classification of the Tetrapods. Volume2, Mammals,
Systematics Association Special Volume 35B, ed. M.
J. Benton. Oxford: Clarendon Press.
Frailey, D. 1978. An early Miocene (Arikareean) fauna
from north central Florida (the SB-1A Local Fauna).
Univ. Kansas, Mus. Nat. Hist., Occ. Pap. 75:1-20.
1979. The large mammals of the Buda Local Fauna
(Arikareean: Alachua County, Florida). Bull. Florida
State Mus., Biol. Sci. 24:123-173.
Frick, C. 1937. Hored ruminants of North America. Bull.
Amer. Mus. Nat. Hist. 69:669 pp.
Ginsburg, L. 1961. La faune des Carnivores Mioc6nes de
Sansan (Gers.). Memoires de la Musdum National
d'Histoire Naturelle (Paris), Series (C) 9:1-190.
and J. Morales. 1992. Contribution a la connaissance
des Mustdlidds (Carnivora, Mammalia) du Miocene
d'Europe: Trochictis et Ischyrictis genres affines et

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

genres nouveaux. Comptes Rendus de I'Academie des
Sciences (Paris), Serie II 315:111-116.
Green, M. 1972. Lagomorpha from the Rosebud Forma-
tion, South Dakota. J. Paleontol. 46:377-385.
__ and J. E. Martin. 1976. Peratherium (Marsupialia:
Didelphidae) from the Oligocene and Miocene of South
Dakota. Pp. 155-168. in Essays on Paleontology in
Honour of Loris Shano Russell, ed. C. S. Churcher.
Athlon. Life Sci. Misc. Publ., Royal Ontario Museum.
Haq, B. U., J. Hardenbol, and P. R. Vail. 1987. Chronology
of fluctuating sea levels since the Triassic. Science
Helbing, H. 1930. Zwei oligocaene Musteliden (Plesictis
genettoides Pomel-Palaeogale angustifrons Pomel.).
Abhandlungen der Schweizeroschen Palaeonto-
logischen Gesellschaft, Band L: 1-36.
1936. Die tertiaren Wirbeltiere des Steinheimer
Beckens. Teil V: Die Carnivoren des Steinheimer
Beckens. A. Mustelidae. Palaeontographica, Supple-
mental Band 8:1-56.
Hodell, D. A., P. A. Mueller, and J. R. Garrido. 1991. Varia-
tions in the strontium isotopic composition of seawater
during the Neogene. Geology 19:24-27.
Hough, J. R. 1948. The auditory region in some members
of the Procyonidae, Canidae, and Ursidae: Its signifi-
cance in the phylogeny of the Carnivora. Bull. Amer.
Mus. Nat. Hist. 92: 67-118.
Hulbert, R.C., Jr. 1992. A checklist of the fossil vertebrates
of Florida. Florida Paleontological Society. Pap. Florida
Paleontol. 6:35 pp.
Hunt, R. M., Jr. 1978. Depositional setting of a Miocene
land mammal assemblage, Sioux County, Nebraska.
Palaeogeo., Palaeoclim., Palaeoecol.24:1-52.
__ 1989. Evolution of the aeluroid Carnivora: Sig-
nificance of the ventral promontorial process of the pet-
rosal, and the origin of the basicranium in the living
families. Amer. Mus. Nov. 2930:1-30.
and R. H. Tedford. 1993. Phylogenetic relation-
ships within the aelurid Carnivora and implications of
their temporal and geographic distribution. Chapter 5,
pp. 53-73 in Mammal Phylogeny (Placentals), ed. F. S.
Szalay, M. J. Novacek, and M. C. McKenna, New York:
Hunter, M. E. 1972. Biostratigraphy and paleontology, in
Oligocene stratigraphy A study of the Lansing Quarry
near Brooksville, Hernando County, Florida. Bay Area
Geol. Soc. First Field Conference Guidebook:11-28.
Jones, D. S., P. A. Mueller, D. A. Hodell, and L. A. Stanley.
1993. 87Sr/86Sr geochronology of Oligocene and Mi-
ocene marine strata in Florida. Pp.15-26 in The Neo-
gene of Florida an adjacent regions, ed. V. A. Zullo, W.
B. Harris, T. M. Scott, and R. W. Portell. Proceedings,
3rd Bald Head Island Conference on Coastal Plains
Geology: Florida Geol. Surv. Spec. Publ. 37:15-26
Korth, W. W. 1992. Fossil small mammals from the Harrison
Formation (late Arikareean: earliest Miocene), Cherry
County, Nebraska. Ann. Carnegie Mus. 61:69-131.

1994. Middle tertiary marsupials (Mammalia) from
North America. J. Paleontol. 68:376-397
1996. A new genus of beaver (Mammalia:
Castoridae: Rodentia) from the Arikareean (Oligocene)
of Montana and its bearing on castorid phylogeny. Ann.
Carnegie Mus. 65:167-179.
___ and R. J. Emry. 1997. The skull ofAnchitheromys
and a new subfamily of beavers (Castoridae, Roden-
tia). J. Paleontol. 71:343-347.
Lange, B. 1970. Mustelictis piveteaui, mustdlidd nouveau
des Phosphorites du Quercy. Annales de Paldontologie
(Vertebrds) 56:73-91.
Lillegraven, J. A., M. C. McKenna, andL. Krishtalka. 1981.
Evolutionary relationships of middle Eocene and
younger species of Centetodon (Mammalia, Insectivora,
Geolabididae) with a description of the dentition of
Ankylodon (Adapisoricidae). Univ. Wyoming Publ.
Macdonald, J. R. 1963. The Miocene faunas from the
Wounded Knee area of western South Dakota. Bull.
Amer. Mus. Nat. Hist. 125:139-238.
1970. Review of the Miocene Wounded Knee fau-
nas of southwestern South Dakota. Bull. Los Angeles
Co. Mus. Nat. Hist. 8:1-82.
Macdonald, L. J. 1972. Monroe Creek (early Miocene)
microfossils from the Wounded Knee area, South Da-
kota. South Dakota Geol. Surv., Rept. Investigations
MacFadden, B. J. 1980. An early Miocene land mammal
(Oreodonta) from a marine limestone in northern
Florida. J. Paleontol. 54:93-101.
and S. D. Webb. 1982. The succession of Miocene
(Arikareean through Hemphillian) terrestrial mammal
localities and faunas in Florida; Pp. 186-199 in Mi-
ocene of the Southern United States, ed. T. Scott and S.
Church. Florida Geol. Spec. Publ. 25.
and R M. Hunt, Jr. 1998. Magnetic polarity stratig-
raphy and correlation of the Arikaree Group, Arikareean
(ate Oligocene-early Miocene) of northwestern Nebraska.
Pp.143-166 in Depositional Environments,
Lithostratigraphy, and Biostratigraphy of the White River
and Arikaree Groups (Late Eocene to Early Miocene,
North America), ed. D. O. Terry, H. E. LaGarry, and R
M. Hunt, Jr.: Geol. Soc. Amer. Spec. Pap. 325.
Martin, L. D. 1987. Beavers from the Harrison Formation
(early Miocene) with a revision of Euhapsis. Pp.73-91 in
Papers in Vertebrate Paleontology in Honor of Morton
Green, ed. J. E. Martin and G. E. Ostrander. Dakoterra,
Matthew, W. D., and W. Granger. 1924. New Carnivora from
the Tertiary of Mongolia. Amer. Mus. Nov. 104:9 p.
Mead, G. A., and D. A. Hodell. 1995. Controls on the
87Sr/86Sr composition of seawater from the middle
Eocene to Oligocene: Hole 689B, Maud Rise, Antarc-
tica. Paleoceanography 10:327-346.
Morgan, G. S. 1989. Miocene vertebrate faunas from the
Suwannee River Basin of north Florida and south

Georgia. Pp. 26-53 in Miocene Paleontology and
Stratigraphy of the Suwannee River Basin of North
Florida and South Georgia, ed. G. S. Morgan. Southeast.
Geol. Soc., 1989 Annual Field Trip Guidebook 30.
1993. Mammalian biochronology and marine-non-
marine correlations in the Neogene of Florida. Pp. 55-
66 in The Neogene of Florida and Adjacent Regions,
eds. V. A. Zullo, W. Burleigh Harris, T. M. Scott, and
R. W. Portell. Florida Geol. Surv. Spec. Publ. 37.
Nowak, R. M. 1999. Walker's Mammals of the World, 6th
edition, Volume 1. Baltimore: The Johns Hopkins
University Press. 836 pp.
Patton, T. H. 1967. Oligocene and Miocene vertebrates from
central Florida. Southeast. Geol. Soc., 13th Field Trip
Guidebook, pp. 3-10.
1969. An Oligocene land vertebrate fauna from
Florida. J. Paleontol. 43:543-546.
Peterson, O. A. 1924. Discovery of fossil mammals in the
Brown's Park Formation of Moffat County, Colorado.
Ann. Carnegie Mus. 15:299-304.
Pratt, A.E. 1989. Taphonomy of the microvertebrate fauna from
the early Miocene Thomas Fann Locality, Florida (U.S.A).
Palaeogeo., Palaeoclim, Palaeoecol. 76:125-151.
Prothero, D. L., and R. J. Emry. 1996. Summary. Pp. 664-
683 in The terrestrial Eocene-Oligocene transition in
North America, ed. D. L. Prothero and R J. Emry. New
York: Cambridge University Press.
and N. Shubin. 1989. The evolution of Oligocene
horses. Chapter 10; pp.142-175 in The Evolution of
Perissodactyls, ed. D. R. Prothero and R. M. Scoch.
New York: Oxford University Press, Oxford Monogr.
Geol. Geophysics 15.
Randazzo, A. F 1997. The sedimentary platform ofFlorida:
Mesozoic to Cenozoic. Pp.39-56 in The Geology of
Florida, ed. A. F. Randazzo and D. S. Jones. Gainesville:
University Presses of Florida.
Rensberger, J. M 1971. Entoptychine pocket gophers (Mamma-
lia, Geomyoidea) of the early Miocene John Day Forma-
tion, Oregon. Univ. California Publ. Geol. Sci. 90:1-209.
1973. Pleurolicine rodents (Geomyoidea) of the
John Day Formation, Oregon, and their relationships
to taxa from the early and middle Miocene, South Da-
kota. Univ. California Publ. Geol. Sci. 102:1-95.
1983. Successions of meniscomyine and allomyine
rodents (Aplodontidae) in the Oligo-Miocene John Day
Formation, Oregon. Univ. California Publ. Geol. Sci.,
124 pp.
Rich, T. H. V. 1981. Origin and history of the Erinaceinae
and Brachyericinae (Mammalia: Insectivora). Univ.
Kansas Mus. Nat. Hist. Occ. Pap. 21:1-116.
__ and T. H. Patton. 1975. First record of a fossil
hedgehog from Florida (Erinaceidae, Mammalia). J.
Mammal. 56:692-696.
and D. L. Rassmussen. 1973. New North Ameri-
can erinaceine hedgehogs (Mammalia: Insectivora).
Univ. Kansas Mus. Nat. Hist. Occ. Pap. 21:1-54.


Schmidt-Kittler, N. 1981. Zur Stammesgeschichte der
marderverwandten Raubtiergruppen (Musteloidea,
Carnivora). Eclogae GeologicaeHelvetiae 74:753-801.
Scott, T. M. 1992. A Geological Overview of Florida.
Florida Geol. Surv. Open File Rept. 50:78 pp.
Simpson, G. G. 1930. Tertiary land mammals of Florida.
Bull. Amer. Mus. Nat. Hist. 59:149-211.
1932. Miocene land mammals from Florida. Florida
State Geol. Surv. Bull. 10:7-42.
1946. Palaeogale and allied early mustelids. Amer.
Mus. Nov. 1320:14 pp.
Stevens, M. S., J. B. Stevens, and M. R. Dawson. 1969. New
early Miocene formation and vertebrate local fauna, Big
Bend National Park, Brewster County, Texas. Texas Me-
morial Museum, Pearce-Sellard Ser., 15:53 pp.
Stirton, R. A. 1935. A review of the Tertiary beavers. Univ.
California Publ. Geol. Sci. 23:391-458.
Stock, C., and E. L. Furlong. 1922. A marsupial from the
John Day Oligocene of Logan Butte, eastern Oregon.
Univ. California Publ. Geol. Sci. 13:311-317.
Tedford, R. H. 1976. Relationships of pinnipeds to other
carnivores (Mammalia). Syst. Zool. 25:363-374.
and D. Frailey. 1976. Review of some Carnivora
(Mammalia) from the Thomas Farm Local Fauna
(Hemingfordian: Gilchrist County, Florida). Amer.
Mus. Nov. 2610:1-9.
___ T. Galusha, M. F Skinner, B. E. Taylor, R. W. Fields,
J. R. Macdonald, J. M. Rensberger, S. D. Webb, and D.
P. Whistler. 1987. Faunal succession andbiochronology
of the Arikareean through Hemplillian interval (late
Oligocene through earliest Pliocene epochs) in North
America. Pp. 153-210 in Cenozoic Mammals of North
America, Geochronolgy and Biostratigraphy, ed. M. O.
Woodburne. Berkeley: University of California Press.
__ and M. E. Hunter. 1984. Miocene marine-nonma-
rine correlations, Atlantic and Gulf Coastal Plains,
North America. Palaeogeo., Palaeoclim., Palaeoecol.
__ J. B. Swinehart, C. C. Swisher III, D. R Prothero,
S. A. King, and T. E. Tierney. 1996. The Whitneyan-
Arikareean Transition in the High Plains. Pp. 312-
334 in The Terrestrial Eocene-Oligocene Transition in
North America, ed. D. R. Prothero and R. J. Emry.
Cambridge: Cambridge University Press.
__ B. E. Taylor, and X. Wang. 1995. Phylogeny of the
Caninae (Carnivora: Canidae): The living taxa. Amer.
Mus. Nov. 3146:1-37.
Van Valen, L. 1966. Deltatheridia, a new order of mam-
mals. Bull. Amer. Mus. Nat. Hist. 132:1-126
Vernon, R. O. 1951. Geology of Citrus and Levy Counties,
Florida. Florida Geol. Surv. Bull. 33:256 pp.
Wang, B. 1992. The Chinese Oligocene: A preliminary
review of mammalian localities and local faunas. Pp.
529-547 in Eocene-Oligocene Climatic and Biotic Evo-
lution, ed. D. R. Prothero and W. A. Berggren.
Princeton NJ: Princeton University Press.

HAYES: Brooksville 2 Local Fauna (Arikareean, Latest Oligocene): Hernando County, Florida

Wang, X. 1994. Phylogenetic systematics of the
Hesperocyoninae (Carnivora: Canidae). Bull Amer Mus
Nat Hist 221:1-207.
andR. H. Tedford. 1996. Canidae. Pp. 433-452 in
The Terrestrial Eocene-Oligocene Transition in North
America, ed. D. R. Prothero and R. J. Emry. New York:
Cambridge University Press.
R. H. Tedford, and B. E. Taylor. 1999. Phyloge-
netic systematics of the Borophaginae (Carnivora:
Canidae). Bull. Amer. Mus. Nat. Hist. 243:1-391.
Webb, S. D., B. MacFadden, and J. A. Baskin. 1981. Geol-
ogy and Paleontology of the Love Bone Bed from the
late Miocene of Florida. Amer. J. Sci. 281:513-544.
White, W. A. 1970. Geomorphology of the Florida Penin-
sula. Florida Geol. Surv. Bull. 51:1-164.
Wilson, R W. 1949. Rodents and lagomorphs of the upper
Sespe. Carnegie Inst. Washington Publ. 584:51-65.
Wolsan, M. 1993. Phylogeny and classification of early
European Mustelida (Mammalia: Carnivora). Acta
Theriologica 38:345-384.
and B. Lange-Badrd. 1996. An arctomorph carni-
voran skull from the Phosphorites du Quercy and the
origin of the procyonids, Acta Paleontologica Polonica
41, 3:277-298
Wood, A. E. 1940. The mammalian fauna of the White

River Oligocene. Part 3. Lagomorpha. Amer. Phil. Soc.
Trans., n. s. 28:271- 362.
Wood, H. E., and A. E. Wood. 1937. Mid-Tertiary verte-
brates from the Texas Coastal Plain; fact and fable.
Amer. Midi. Nat. 18:129-146.
Woodburne, M. O., and C. A. Swisher, III. 1995. Land
mammal high-resolution geochronology, intercontinen-
tal overland dispersals, sea level, climate, and
vicariance. Pp. 335-364 in Geochronology, Time Scales,
and Global Stratigraphic Correlation, ed. W. A.
Berggren, D. V. Kent, M-P. Aubry, J. Hardenbol. Soc.
Sediment. Geol., Spec. Publ. 54.
Xu, X. 1994. Evolution of Chinese Castoridae. Pp. 77-98
in Rodent and Lagomorph Families of Asian Origin
and Diversification, ed. Y. Tomida, L. Chuankuei, and
T. Setoguchi. Tokyo: Proceedings of Workshop WC-2,
29th International Geological Congress, Kyoto, Japa-
nese National Science Museum.
1996. Castoridae. Chapter 20; Pp. 417-432 in The
Terrestrial Eocene-Oligocene Transition in North
America, ed. D. R Prothero and R. J. Emry. Cambridge:
Cambridge University Press.
Yon, J. W., and C. W. Hendry. 1972. Suwannee Limestone
in Hernando and Pasco Counties, Florida. Florida Bur.
Geol. Bull.54:1-42.

The BULLETIN OF THE FLORIDA MUSEUM OF NATURAL HISTORY publishes original biological research. Manuscripts
dealing with natural history or systematic problems involving the southeastern United States or the neotropics are especially
welcome. Submitted papers should be of medium length, ca. 10,000-60,000 words. Authors should include the names of
three suggested reviewers. The BULLETIN receives manuscripts in confidence and protects the confidentiality of the content.
The BULLETIN is distributed worldwide through institutional exchanges and standing orders. Fifty free copies are
sent to the first author who may order additional separates at cost (see below).

Format- The BULLETIN encourages the following format. Contributors also should consult recent numbers of the BULLETIN
for style and format, as well as the book Scientific Style and Format, CBE Style Manual for Authors, Editors, and Publishers,
6th Edition, 1994 (published by the Council of Biology Editors). Authors should submit three hard copies of the manuscript
with complete tables, table legends, figures (one set original, two photocopies), and figure legends, plus a floppy disc or CD
compatible with Word for Windows. Manuscripts must be typed, one side only, on 8 1/2" X 11" white bond paper, and
double-spaced throughout. All pages must be numbered consecutively, including references, figure legends, footnotes, and
tables. All margins should be at least 1 inch (25 mm) wide, and not justified on the right margin. Manuscripts not submitted
in the format will be returned to the authors.
In most cases the parts of the manuscript should be as follows.
a. Title page-A separate sheet with the title, name(s) and addresses) of authorss, and a suggested running title.
b. Abstract-A separate sheet with 200 words or less summarizing the paper, and listing five boldfaced key words.
c. Conventions-Introduction, material and methods, results, discussion, and acknowledgments. Taxonomic reviews: the
first mention in the text of a binomial species name should include the taxonomic authority and the year of publication [e.g.,
Notogillia wetherbyi (Dall, 1865)]. Thereafter the full generic name and species epithet must be written out each time the
name first appears in a paragraph. The generic name may be abbreviated in the remainder of that paragraph as follows: N.
wetherbyi. The reference need not be cited when author and date are given only as authority for a taxonomic name. Use
italics for species names and metric units for measurements. A figure is cited in the text as Fig. (e.g., Fig. 3). Authors should
indicate the placement of figures in the left margin of the text.
d. Literature cited-References in the text should give the name of the authors) followed by the date of publication: for one
author (Smith, 1999), for two authors (Schultz and Whitacre, 1999), and for more than two (Britt et al., 1983). Only papers
cited in the text may be included. All authors' last names are given in full. Initials are used for first and middle names. Each
citation must be complete, with all journal titles unabbreviated, and in the following forms.
Eisenberg, J. F. and J. R. Polisar. 1999. The mammal species of north-central Venezuela. Bulletin of the Florida Museum
of Natural History, 42 (3):115-160.
Lundell, C. L. 1937. The Vegetation of Petdn. Carnegie Institute of Washington, Publication 478. Washington, D.C.
Composite Works:
Gardner, A. L. 1993. Didelphimorpha. Pp. 15-24 in D. Wilson and D. M. Reeder, eds. Mammal Species of the World.
Washington, D.C.: Smithsonian Institution Press, 1206 pp.
e. Legends-Legends should be typed on a separate sheet for each table and each set of figures. Abbreviations used in the
figures should be included in the figure legends.
f. Tables-Each table should be on a separate sheet. Tables must be numbered with Arabic numerals. Each table should be
headed by a brief legend. Avoid vertical rules.
g. Figures-All illustrations are referred to as figures. Authors should submit one set of original figures consecutively numbered
with Arabic numerals. Clear xerographic copies are suitable for reviewer's copies of submitted manuscript. All figures must
be ready for publication and comply with the following standards.
Photographs and halftone reproductions must be on glossy high-contrast paper, grouped as appropriate, and in
focus with good contrast. A scale bar may be used in the photograph; otherwise, the figure legend should give the size. If the
background of photographs (especially those of specimens) is not desired, amberlith may be cut to mask out the background.
Text figures should be in black ink and completely lettered. Drawings should be made with dense black waterproof
ink on quality paper or illustration board. All lettering must be medium weight (bold for small type white on black), sans-
serif in cutout, dry transfer, or lettering guide letters. Lettering must be not less than 2 mm high or greater than 5 mm high
after reduction for publication.
The maximum size for figures is 9" x 14". Figures should be approximately typepage width (6-1/2") or column
width (3-1/s"), and should comply with page format. Figures should be labeled on the back to indicate the top of the figure and
to identify author's name, manuscript title, and figure number.
Proofs-Page proofs will be sent to the first author to check for printer's errors and returned to the Editor promptly.
Page charges-Publication necessitates page charges to the first author, which are determined after consultation with the first
author upon final acceptance of the manuscript. The first author is responsible for any charges incurred for alterations made
by him on galley or page proofs, other than corrected printer's errors. The MUSEUM will send an invoice to the first author
for all charges upon completion of publication. Color illustrations will be published at extra cost to the authorss.
Reprints-Order forms are sent to the first author with the page proofs. The first author is responsible for placing all orders.


University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs