2013 SOUTHEASTERN NATURALIST 12(2):427–438
Extinct Pterygoboa (Boidae, Erycinae) from the Latest
Oligocene and Early Miocene of Florida
Jim I. Mead1,* and Blaine W. Schubert1
Abstract - Members of the Erycinae are small to medium-sized, semi-fossorial snakes in
the family Boidae (Squamata, Serpentes) known today from Africa, southwestern Asia,
India, and western North America. Erycines were the predominate snakes in faunas of
North America during the Paleogene. In North America, only the minute, extinct erycine
Pterygoboa is known to have additional wing-like processes situated on the postzygapophyses
of the vertebrae. Here we report on Pterygoboa from one latest Oligocene
(25–24 Ma) (White Springs) and two early Miocene (≈20–19 Ma) localities (Miller,
Thomas Farm) in Florida. These specimens represent a significant chronological and
southern geographic range extension for the genus and permit an amendment to the morphological
description of this unusual snake.
Introduction
Members of the monophyletic subfamily Erycinae (sand-snakes) are small to
medium-sized, semi-fossorial snakes (Squamata, Serpentes) in the family Boidae
(Macrostomata, Booidea). Today members of the subfamily live in Africa, southwestern
Asia, India, and western North America (Charina, Lichanura). During
the Paleogene (early portion of the Cenozoic; Paleocene to Oligocene), erycine
snakes were the predominate snakes in faunas of North America (Holman 2000).
Many fossil specimens in North America are attributed to the Erycinae; however,
Szyndlar and Rage (2003) state that such referral to this subfamily is “… groundless
and the fossils in question actually may have represented other groups of
the Booidea.” Their statement aside, a number of North American researchers
believe that during the latest Oligocene and early Miocene (Arikareean North
American Land Mammal Age [NALMA]) the abundance and diversity of erycine
snake species began to wane at about the same time that the colubroid snakes
began to expand (Holman 2000, Parmley and Holman 1995). Here we report on
an unusual and minute, extinct erycine snake, Pterygoboa, recently recognized
in two early Miocene (Hemingfordian NALMA) localities (Miller local fauna,
Thomas Farm) and one latest Oligocene (middle Arikareean NALMA) locality
(White Springs) in northern Florida.
Erycine snakes
Overview of erycine. Erycine vertebrae differ from all other snakes in having
additional apophyses (small, bony outgrowths), known as pterapophyses, located
on the neural arch between the neural spine and the postzygapophyses of the
1Department of Geosciences and Don Sundquist Center of Excellence in Paleontology,
East Tennessee State University, Johnson City, TN 37614. *Corresponding author
-MEAD@mail.etsu.edu.
428 Southeastern Naturalist Vol. 12, No. 2
caudal vertebrae (Fig. 2: pt; Hoffstetter and Rage 1972, Szyndlar 1994, Szyndlar
and Schleich 1994). Trunk vertebrae within the clade (which characteristically
lack these additional apophyses) differ little from one another and therefore
create taxonomic and identification issues. European and North American researchers
habitually differ in their approach to identifying fossil erycine snake
vertebrae; “fossil trunk vertebrae should not be interpreted as belonging to the
Erycinae unless they are accompanied by caudal vertebrae displaying complex
morphology characteristic” of the subfamily (Szyndlar and Rage 2003). North
American Neogene erycine species are diagnosed based on trunk vertebrae (see
discussion and views in Szyndlar and Rage 2003). In the view of these authors,
the extinct Calamagras (including the possibly congeneric Ogmophis) and the
extant Charina and Lichanura are considered the only true erycines in North
America. Holman (1976a) assigned the extinct Pterygoboa to Erycinae; however,
Szyndlar and Rage (2003) disagreed because “additional vertebral apophyses in
[Pterygoboa] are not restricted to the caudal portion of the column … which cast
doubts on [its] assignment” to the subfamily (see amendment below).
Pterygoboa miocenica Holman is diagnosed, in part, as “trunk and caudal
vertebrae with winglike elaborations of the postzygapophyses”. In naming the
taxon, and again in 1977, Holman referenced these wing-like apophyses (Holman
1976a, 1977). For whatever reason, in his description of a second species,
P. delawarensis Holman, he termed the diagnostic wing-like structures as “pterapophyses”—“…
pterapophyses present on postzygapophysis; …”; this is an error
in terminology. Hoffstetter and Rage (1972) used the term “pterapophyses” for
the supplementary apophyses located on the neural arch between the neural spine
and the postzygapophyses; Sood (1941) used the term “accessory lateral process”
for these same apophyses. Pterapophyses on the neural arch are not equivalent to
the wing-like structures observed on the lateral edges of the postzygapophyses as
described by Holman (1976a, 1977, 1998) to diagnose the genus Pterygoboa and
in helping to diagnose the two species. Rage (1984) referred to the wing structures
as “additional processes above the postzygapophyses.” Szyndlar (1994)
named the high wing-like process above the postzygapophysis on caudal vertebrae
of the European Eryx a “postzygapophyseal wing” (see also Szyndlar and
Schleich 1994); this nomenclature will be followed here. We suspect that these
postzygapophyseal wings are not homologous to Soods’s (1941:figs. 2–4) “winglike
plate (alar plate), extending upwards and backwards from the middle of the
zygapophysial ridge.” Here we state that the postzygapophyseal wings used by
Szyndlar (1994) for caudal vertebrae is homologous to the wing-like elaborations
above the postzygapophyses observed on trunk vertebrae of Pterygoboa.
We also recognize on Pterygoboa the apophysis termed pterapophysis and use
the definition of Hoffstetter and Rage (1972). The revised osteological terms as
described above are used below and in place of Holman’s (1976a, 1977, 1998,
2000) original terminology.
General morphology of species. There are two species described in Pterygoboa,
with P. miocenica the genotype (Holman 1976a) and secondarily
2013 J.I. Mead and B.W. Schubert 429
P. delawarensis (Holman 1998) (Fig. 1). Although the genus is distinct among
the booids, the two species seem near-identical in morphological description
(see below). A large number of vertebrae are assigned to P. miocenica, yet only
a single trunk vertebra is known for P. delawarensis. Original diagnoses and
descriptions of the species are provided in citations above; revised diagnoses
and descriptions for both species are provided by Holman (2000) along with
amended osteological terminology discussed below.
Pterygoboa miocenica is known from the type locality in the Black Bear
Quarry II of the Rosebud Formation, Bennett County, SD (Holman 1976a;
Fig. 1), and is of the earliest Hemingfordian NALMA (approximately 19 Ma;
Tedford et al. 2004). The snake is also recovered from the Myers Farm fauna,
Webster County, NE (Fig. 1) which is middle Barstovian NALMA (approximately
13.5 Ma; Holman 1977, Tedford et al. 2004). Holman (2000) found in
analyzing the 76+ trunk and 28 caudal vertebrae from the Myers Farm locality
that there is a lot of variation in the overall extent of the development of the
postzygapophyseal wing, which he related to intracolumnar variation. Morphological
information comes from the revised description of the type, paratype,
Figure 1. Map of localities containing Pterygoboa. Specimens identified as P. miocenica:
BBQ II (open circle), type locality at Black Bear Quarry II, Rosebud Formation, early
Hemingfordian LMA, South Dakota; MF (open circle), Myers Farm Site, middle Barstovian
NALMA, Nebraska. Pterygoboa from the Miller local fauna (black square), earliest
Hemingfordian NALMA, Florida; Thomas Farm (open triangle) early Hemingfordian
NALMA, Florida; White Springs local fauna (asterisk), middle Arikareean NALMA,
Florida. The single locality of P. delawarensis: PF (black dot), type locality of at Pollack
Farm site, early Hemingfordian NALMA, Delaware.
430 Southeastern Naturalist Vol. 12, No. 2
and other vertebrae assigned to the species, and is based on a minimum of 4
cervical, 79 trunk, and 29 caudal vertebrae. Morphological characters of Holman
(2000) include: 1) vertebrae are “much” wider than long, 2) neural arch is
moderately depressed (flattened), 3) postzygapophyseal wing is approximately
two-thirds the length of the centrum, 4) anterior and posterior ends of postzygapophyseal
wings are squared-off to blunt-pointed, 5) synapophyses are slightly
divided into dia- and parapophyses but not distinct from one another, 6) deep
subcentral grooves are present, and 7) neural spine is higher than it is long
along the base, yet there is variation from moderately high to low as shown in
Rage (1984), and it is typically swollen dorsally (Holman 1998, 2000).
Pterygoboa delawarensis is known only from the Pollack Farm locality,
Calvert Formation, Kent County, DE (Fig. 1), and is of the early Hemingfordian
NALMA (approximately 18 Ma; Benson 1998). The description is based on a
single trunk vertebra (Holman 1998). The following morphological information
comes from the revised description of the type and includes: 1) vertebra is slightly
wider than long, 2) neural arch is highly depressed (flattened), 3) postzygapophyseal
wing relatively short compared to length of centrum (based on illustration of
holotype (Holman 1998:fig. 1), 4) postzygapophyseal wings are pointed at both
anterior and posterior ends, 5) it is unknown if the synapophyses are divided into
dia- and parapophyses, 6) distinct subcentral ridges and grooves present, and
7) the neural spine is low and dorsally swollen (Holman 1998, 2 000).
Study Area
Faunal remains from Thomas Farm (TF; one of the most systematically excavated
and analyzed fossil localities in the southeast US; Fig. 1) indicate that
the fossils belong to the early Hemingfordian NALMA, dating about 19–18 Ma
(specifically sub-event He1; R. Hulbert, Florida Museum of Natural History,
Gainsville, FL, February 2012 pers. comm.). Located in Gilchrist County, the deposit
is localized, and most of the layers cannot be traced laterally over a distance
greater than 20 m (Pratt 1990). Geologic features preserved at the locality imply
that the site was a large sinkhole (see Pratt 1990:fig. 15), with the recovered fauna
representing an autochthonous assemblage. Faunal remains entered the sinkhole
1) as victims of a natural trap scenario, 2) as skeletal elements that accumulated
on the surface surrounding the sinkhole and subsequently washed in, 3) as skeletal
remains on talus slopes along the walls of the sinkhole such as from roosting
behavior, and/or 4) as skeletal remains from animals that inhabited a cave (Morgan
and Czaplewski 2003, Pratt 1990). Modification of bone by agents such as
weathering or carnivores was not a significant factor (Pratt 1990). The amount, if
any, of raptor activity and how it may have influenced the microfaunal species in
the deposit is not known at this time. Pratt (1990) suggests that the environment
around the sinkhole was decidedly tropical and forested rather than open terrain.
The Miller Local Fauna (MLF) contains an extensive diversity of vertebrates,
but to date most fossil groups have not been studied in any detail. Specimens
2013 J.I. Mead and B.W. Schubert 431
were collected from the Suwannee River in Dixie County, FL (Fig. 1). Recovered
borophagine canid, hesperocyonine canid, and undescribed mustelids support an
age assignment of early Hemingfordian NALMA (≈20 Ma, about 1–2 million
years older than TF; Baskin 2003; R. Hulbert, May 2011 pers. comm.).
The White Springs 3B local fauna (WS) occurs in sediments of the Parachucla
Formation exposed along the banks of the Suwannee River near the town of
White Springs, Columbia County, FL (Fig. 1.). The local fauna contains nearshore
marine and terrestrial vertebrates (MacFadden and Morgan 2003, Morgan
1989). Biochronological placement here follows Albright et al. (2008) and Tedford
et al. (2004) in placing WS in middle Arikareean (Ar3) NALMA of the latest
Oligocene, 25–24 Ma.
Methods
All Florida specimens are cataloged into the collections of the Florida
Museum of Natural History, University of Florida, Gainesville (UF). Eight
vertebrae were recovered from MLF: 3 nearly complete trunk (UF 268750,
268751, 272799), 2 complete caudal (UF 268754, 272797), and 3 fragmented
trunk (UF 268752, 268753, 272798). Six trunk vertebrae were recovered from
TF (UF 268702-268705, 271969) and 1 juvenile caudal (UF 271968). Seventeen
trunk vertebrae (of a single individual; UF 125435) were recovered from
WS. Terminology and measurements of vertebrae follow Auffenberg (1963),
Hoffstetter and Gasc (1969), LaDuke (1991), and Szyndlar (1994). Only a few
of the vertebrae listed above will be described in detail.
Results
MLF trunk vertebra UF 272799 (Fig. 2A, B), the most complete trunk vertebra
described here, is 2.2 mm in centrum length. The neural arch width is 2.1 mm,
and is highly flattened in anterior view and slightly arched in posterior view. The
neural spine is complete (1.5 mm long), showing a slightly swollen and flattened
crest (but not to the extent of a neural plate [Sood 1941] or as observed on many
snakes species; e.g., see Bogert 1964, Szyndlar and Rage 2003) and overhangs
posterior to the neural arch. Immediately below the neural spine apex and separate
from the crest are two accessory spines that develop at the posterior end of
the neural spine and run the length of the crest, diverging slightly laterally (Fig.
2A, B). Postzygapophyseal wings are distinct, sweeping dorsally and form ridges
that are square to blunt-pointed at both ends, 1.3 mm long. Prezygapophysis
processes are extremely reduced to protruding barely beyond the edge of the
prezygapophysis. The haemal keel is slight and minutely projecting; there are
two subcentral foramina. Two pterapophyses are observed.
MLF trunk vertebra UF 268750 is 2.1 mm in centrum length. The neural
arch width is 1.9 mm. The neural arch is highly flattened in anterior view
and only slightly arched in posterior view. The neural spine is damaged,
which allows no information of its structure. The neural arch posterior
432 Southeastern Naturalist Vol. 12, No. 2
Figure 2. Trunk and caudal vertebrae identified as Pterygoboa from Florida. A) dorsal
and B) anterior: Miller Local Fauna, trunk vertebra, UF 272799. C) dorsal and D) anterior:
Thomas Farm, trunk vertebra, UF 271969. E) dorsal and F) anterior: Miller Local
Fauna, caudal vertebra, UF 268754. G) dorsal and H) anterior: Miller Local Fauna, caudal
vertebra, UF 272797. Abbreviations: h = haemapophysis, np = neural plate at crest
of neural spine, ns = neural spline, pl = pleurapophysis as blade, po = postzygapophysis,
pow = postzygapophysis wing, pt = pterapophysis, prp = prezygapophysis process. Scale
bars equal 1 mm; A–F same scale; G–H same scale.
2013 J.I. Mead and B.W. Schubert 433
notch is U-shaped as viewed dorsally. A postzygapophyseal wing on the
postzygapophysis is a dorsally oriented blade, straight-topped and 1.1 mm long
with square-blunt ends. The prezygapophyses are slightly dorsally oriented,
with the surface obovate in shape. There is slight evidence of a prezygapophysis
process. Synapophyses (paradiapophyses) are present but damaged; thus, no
indication of a parapophysis projection. The haemel keel is an arched, slightly
projecting keel oriented the full length of the centrum. There are two subcentral
foramina in addition to two minute pterapophyses positioned closer to the
postzygapophyses than to the neural spine.
MLF trunk vertebra UF 268751 has a centrum length of 1.8 mm; the neural
arch width is 1.7 mm. The neural spine is flattened in anterior view and slightly
arched in posterior view. The neural spine is ≈0.7 mm high, slightly swollen at the
crest, and does not protrude (overhang) posteriorly to the neural arch. The neural
arch posterior notch is U-shaped as viewed dorsally. The neural spine is 0.8 mm
long at the base (45% of the centrum length) and low in height, with evidence
of slight swelling at the crest. The prezygapophysis is slightly dorsally oriented
with an obovate shape. There is no evidence of a prezygapophysis process. The
postzygapophyseal wing is dorsally oriented, straight-topped, and 0.9 mm long
with blunt-pointed ends. The synapophysis is present with slight evidence of a
division into a diapophysis and a parapophysis; there is a slight parapophysis projection.
A paracotylar notch is present, but there is no evidence of a paracotylar
foramen. The haemel keel is arched, slightly projecting, and oriented most the
length of the centrum length. There are two pin-point subcentral foramina. There
is no evidence of pterapophyses, which we assume may indicate that the vertebra
is more anteriorly positioned along the vertebral column.
MLF caudal (postcloacal) vertebra UF 268754 (Fig. 2E, F) has a centrum
length of 1.8 mm and a neural arch width of 1.8 mm. The neural arch is highly
flattened in anterior view and only slightly arched in posterior view. The neural
spine length is 1.3 mm, or about 72% the centrum length, and is situated along the
posterior portion of the neural arch. The neural spine is relatively low, expanded
dorsally, highly bifurcated at the apex, and has minute accessory apophyses
immediately ventral to the apex (Fig. 2E, F). The prezygapophysis is dorsally
directed with a minute prezygapophysis process that projects anteriorly. The
postzygapophysis wing combines with the postzygapophysis to form a dorsally
directed blade that is 1.2 mm long (67% of the centrum length) and blunt-pointed
at both ends. Pleurapophyses occur ventral to prezygapophyses and are large
expansions that are laterally directed. The area of the haemal keel is wide, short,
mostly flattened, and shows incipient morphing into a haemapophysis. Pterapophyses
are situated between the neural spine and postzygapophyseal wings
and are oriented toward the anterior.
MLF caudal vertebra UF 272797 (Fig. 2G, H) has a centrum length of 1.4
mm and a neural arch width of 1.4 mm. The vertebra is slightly incomplete due
to etching along the anterior edge of the neural arch and zygosphene. The neural
spine occupies most of the length of the neural arch and splays out at the apex
434 Southeastern Naturalist Vol. 12, No. 2
forming a neural plate (Sood 1941) with a shallow groove running the length. A
V-notch occurs at both ends of the neural plate. Postzygapophyseal wings are 1.1
mm long with blunt-pointed ends. Two paracotylar foramina are present. Pleurapophyses
form blades. The prezygapophysis process is positioned well ventral
to the angled prezygapophyses. The hypapophysis is bulbous at the apex, with
incipient haemapophyses forming.
Trunk vertebra UF 271969 (Fig. 2C, D) and others from TF have centrum
lengths ranging from 1.8 to 2.2 mm; neural arch widths range from 1.8 to 2.2
mm. All vertebrae from TF show relatively short postzygapophysis wings, with
the one on UF 268705 being complete and 0.5 mm long. Those on UF 271969 are
incomplete but are approximately 1.0 mm long and have one complete anterior
end and one complete posterior end with both ends blunt-pointed. Two vertebrae
(UF 268703, 268704) have pterapophyses on the neural arch between the neural
spine and the postzygapophysis wings indicating that the vertebrae are probably
more caudally positioned. UF 271968 is a juvenile caudal vertebra (based on
size of the neural canal relative to the rest of the vertebra) with small but clearly
developed postzygapophysis wings.
The 17 trunk vertebrae recovered from WS are highly fragmented but are all
about the same overall size. Characters of the vertebrae and the pattern to the
breakage imply that the vertebrae are from the same individual. All vertebrae
show some preservation of a postzygapophyseal wing. Two vertebrae have a relatively
high neural spine that occupies the posterior portion of the neural arch; one
has the conspicuous swollen crest. Pterapophyses occur on some of the vertebrae
along the neural arch.
Discussion and Conclusions
Here we present the first records of the erycine boid Pterygoboa from the
North American Gulf Coast region (Auffenberg 1963, Holman 2000). Previous
records indicate that the genus is known from the early Miocene (early Hemingfordian
NALMA and middle Barstovian NALMA; ≈20.0 Ma–13.5 Ma) fossil
deposits in South Dakota, Nebraska, and Delaware (Fig. 1). This distribution
makes the MLF and TF records in Florida presented here of biogeographic importance
in showing a southern distribution at the same time the taxon occurred
further north and west. Vertebrae from WS indicate that Pterygoboa was known
in at least the Florida region as early as the middle Arikareean NALMA, latest
Oligocene, which extends the earliest age for the taxon back another 4–5 million
years to about 24–25 Ma. These three new records from Florida illustrate that the
geographic distribution and chronological extent of Pterygoboa was far greater,
more southern in extent, and older than previously known (Fig. 1).
The morphological characters observed on the 8 vertebrae from the MLF,
the 7 vertebrae from TF, and the 17 vertebrae from WS exhibit many characters
found only on Pterygoboa miocenica and P. delawarensis. Vertebrae from
MLF, TF, and WS are characteristically longer than wide with only about 0.1
2013 J.I. Mead and B.W. Schubert 435
mm difference; whereas, most Pterygoboa miocenica vertebrae and the single
vertebra from P. delawarensis are slightly wider than long. MLF, TF, and WS
vertebrae all exhibit a neural arch that is highly to well-flattened in anterior
view; this characteristic is similar to both P. miocenica and P. delawarensis. The
neural spine, when preserved, is typically low with some evidence of swelling
at the crest to the extent of developing a flattened neural plate; this morphology
is similar to the observed range on both P. miocenica and P. delawarensis. The
postzygapophyseal wings on MLF, TF, and WS vertebrae are not all preserved
but those that are have more blunt-pointed to square-blunt anterior and posterior
ends. This character is one of the major distinctions noted between P. miocenica,
with its blunt-ended postzygapophyseal wings, and, P. delawarensis with its
pointed ends (Holman 1998). In contrast, Rage (1984) illustrates a trunk vertebra
of P. miocenica from the Myers Farm Site that does exhibit more pointed ends
to the postzygapophyseal wings. We infer that this character of blunt ends versus
pointed ends to the postzygapophyseal wings is variable (as stated in Holman
2000) and in likelihood should not be used to distinguish between the two species.
The postzygapophyseal wings on Pterygoboa trunk vertebrae are variable
in length as compared to the centrum length, which we infer depends on the position
within the vertebral column. Those that are more mid-trunk length (MLF UF
268751, P. delawarensis holotype, Myers Site P. miocenica) are typically shorter.
The postzygapophyseal wings on more caudally situated and caudal vertebrae
exhibit wider, more robust, longer, and more blunt-pointed ends.
Three (including the caudal vertebra) of the eight vertebrae from the MLF,
two from TF, and two from WS exhibit pterapophyses. We infer that the presence
of these represent trunk vertebrae that are more caudally positioned and not
mid-trunk or cervical in position. These pterapophyses have not been mentioned
in previous descriptions of Pterygoboa (Holman 1976a, 1998, 2000; Rage 1984;
although figured by Holman 1998:fig. 1A) and clearly demonstrate that Pterygoboa
is indeed a member of the Erycinae (sensu Szyndlar and Rage 2003).
The morphology of the type, P. miocenica, is stated to be clearly understood
(Holman 2000), with variations fairly well delineated based on many specimens
from the type site. However, only a single specimen is known for P. delawarensis.
In our opinion, its morphological characters seem to blend into the
amount of variation observed on P. miocenica rendering the specific status of
P. delawarensis questionable. Consequently we feel that Pterygoboa specimens
described here from Florida could either represent a large geographic and/or
temporal range extension for P. miocenica, or, a separate species of Pterygoboa
with similar vertebral morphology. Future recovery of cranial elements might
help resolve this issue.
Pterygoboa was found associated with other early Hemingfordian-age extinct
erycine boids (Ogmophis miocampactus Holman, Calamagras weigeli
Holman, Charina prebottae Brattstrom) and extinct colubrines at the Black
Bear Quarry II type locality (Brattstrom 1958; Holman 1972, 1976a). At the
Pollack Farm locality of equivalent age, Pterygoboa was recovered with
436 Southeastern Naturalist Vol. 12, No. 2
the erycine Calamagras, colubrines (Ameiseophis robinsoni Holman, Pollackophis
depressus Holman), a viperid (Viperidae), along with the anguid
lizard (Ophisaurus) and a possible Crocodylus (crocodilian) (Holman 1976c,
1998). The snake fauna from the MLF is just now being analyzed and nothing
is reconstructed regarding the early Hemingfordian environmental setting. Of
near-equivalent age to MLF is the well-known TF locality, an extremely rich
and diverse vertebrate assemblage. Snakes from TF include the erycine (Calamagras
floridanus Auffenberg), boine (Pseudoepicrates stanolseni Vanzolini),
and colubrines (Paraoxybelis floridanus Auffenberg, Pseudocemophora antiqua
Auffenberg) (Auffenberg 1963, Vanzolini 1952). With the report here,
Pterygoboa is now known from TF. By the middle Barstovian NALMA at
Myers Farm locality, Pterygoboa was found with fewer other erycine snakes
(Geringophis depressus Holman) but with multiple species of colubrids,
elapids (Elapidae), and possibly a viperid (Holman 1976b, see list in Holman
1977). Clearly much remains to be studied from the MLF, TF, and WS localities
as well as other well-known fossil sites where the minute squamates have
not received sufficient analytical attention.
Acknowledgments
We thank Andreas Kerner for bringing the Miller local faunal fossil remains to our
attention and their donation to the Florida Museum of Natural History. Andreas Kerner,
Sandra L. Swift, and Kevin Chovanec are greatly appreciated for sorting bones from sediments.
Richard Hulbert is thanked for the loan of the fossils from the Florida Museum of
Natural History and discussions about the Thomas Farm locality. We greatly appreciate
Sandra L. Swift for making the illustrations. Partial funding support for this project was
received from the Department of Geosciences and the Don Sundquist Center of Excellence
in Paleontology, East Tennessee State University. Zbigniew Szyndlar and Dennis
Parmley facilitated immensely in helping us to fully understand the complexities of erycine
osteology and for edits to our manuscript.
Literature Cited
Albright, L.B., M.O.Woodburne, T.J. Fremd, C.C. Swisher, B.J. MacFadden, and G.R.
Scott. 2008. Revised chronostratigraphy and biostratigraphy of the John Dave Formation
(Turtle Cove and Kimberly Members), Oregon, with implications for updated
calibration of the Arikareean North American Land Mammal Age. Journal of Geology
116:211–237.
Auffenberg, W. 1963. The fossil snakes of Florida. Tulane Studies in Zoology
10:131–216.
Baskin, J.A. 2003. New procyonines from the Hemingfordian and Barstovian of the Gulf
Coast and Nevada, including the first fossil record of the Potosini. Bulletin of the
American Museum of Natural History 279:125–146.
Benson, R.N. 1998. Geology and Paleontology of the Lower Miocene Pollack Farm Fossil
Site, Delaware. Delaware Geological Survey, Special Publication 21:1–191.
Bogert, C.M. 1964. Snakes of the genera Diaphorolepis and Synophis and the colubroid
subfamily Xenoderminae (Reptilia, Colubridae). Senckenbergiana Biologica
45:509–531.
2013 J.I. Mead and B.W. Schubert 437
Brattstrom, B.H. 1958. New records of Cenozoic amphibians and reptiles from California.
Bulletin of the Southern California Academy of Sciences 57:5–12.
Hoffstetter, R., and J.-P. Gasc. 1969. Vertebrae and ribs of modern reptiles. Pp. 201–310,
In C. Gans (Ed.). Biology of the Reptilia. Morphology A. Academic Press, London,
UK. 373 pp.
Hoffstetter, R., and J.-C. Rage. 1972. Les Erycinae fossils de France (Serpentes, Boidae)
comprehension et histoire de la sous-famille. Annales de Paléontologie (Vertébrés)
58:81–124.
Holman, J.A. 1972. Herpetofauna of the Calf Creek local fauna (Lower Oligocene:
Cypress Hills formation) of Saskatchewan. Canadian Journal of Earth Sciences
9:1612–1631.
Holman, J.A. 1976a. Snakes from the Rosebud Formation (middle Miocene) of South
Dakota. Herpetologica 32:41–48.
Holman, J.A. 1976b. Snakes of the Gering formation (Lower Miocene) of Nebraska.
Herpetologica 32:88–94.
Holman, J.A. 1976c. Snakes of the Split Rock Formation (Middle Miocene), Central
Wyoming. Herpetologica 32:419–426.
Holman, J.A. 1977. Upper Miocene snakes (Reptilia, Serpentes) from southeastern Nebraska.
Journal of Herpetology 11:323–335.
Holman, J.A. 1998. Reptiles of the lower Miocene (Hemingfordian) Pollack Farm Fossil
Site, Delaware. Pp. 141–148, In R.N. Benson (Ed.). Geology and paleontology of
the lower Miocene Pollack Farm Fossil Site, Delaware. Delaware Geological Survey,
University of Delaware, Newark, DE. Special Publication 21.
Holman, J.A. 2000. Fossil Snakes of North America. Indiana University Press, Bloomington,
IN. 357 pp.
LaDuke, T.C. 1991. The fossil snakes of Pit 91, Ranch La Brea, California. Natural History
Museum of Los Angeles County, Contributions in Science 424:1–28.
MacFadden, B.J., and G.S. Morgan. 2003. New oreodont (Mammalia, Artiodactyla) from
the late Oligocene (early Arikareean) of Florida. Bulletin of the American Museum of
Natural History 279:368–396.
Morgan, G.S. 1989. Miocene vertebrate faunas from the Suwannee River basin of north
Florida and south Georgia. Pp. 26–53, In G.S. Morgan (Ed.). Miocene paleontology
and stratigraphy of the Suwannee River basin of north Florida and south Georgia.
Southeastern Geological Society, Gainesville, FL, Guidebook 30.
Morgan, G.S., and N.J. Czaplewski. 2003. A new bat (Chiroptera: Natalidae) from the
early Miocene of Florida, with comments on natalid phylogeny. Journal of Mammalogy
84:729–752.
Parmley, D., and J.A. Holman. 1995. Hemphillian (late Miocene) snakes from Nebraska,
with comments on Arikareean through Blancan snakes of midcontinental North
America. Journal of Vertebrate Paleontology 15:79–95.
Pratt, A.E. 1990. Taphonomy of the large vertebrate fauna from the Thomas Farm locality
(Miocene, Hemingfordian), Gilchrist County, Florida. Bulletin of the Florida Museum
of Natural History 35:35–130.
Rage, J.-C. 1984. Serpentes. Encyclopedia of Paleoherpetology 1 1:1–80.
Sood, M.S. 1941. The caudal vertebrae of Eryx johnii (Russell). Proceedings of the Indian
Academy of Sciences 14:390–394.
438 Southeastern Naturalist Vol. 12, No. 2
Szyndlar, Z. 1994. Oligocene snakes of southern Germany. Journal of Vertebrate Paleontology
14:24–37.
Szyndlar, Z., and J.-C. Rage. 2003. Non-erycine Booidea from the Oligocene and Miocene
of Europe. Institute of Systematics and Evolution of Animals, Polish Academy
of Sciences, Kraków, Poland.
Szyndlar, Z., and H.-H. Schleich. 1994. Two species of the genus Eryx (Serpentes; Boidae;
Erycinae) from the Spanish Neogene with comments on the past distribution of
the genus in Europe. Amphibia-Reptilia 15:233–248.
Tedford, R.H., L.B. Albright, A.D. Barnosky, I. Ferrusquia-Villafranca, R.M. Hunt, J.E.
Storer, C.C. Swisher, M.R. Voorhies, S.D. Webb, and D.P. Whistler. 2004. Mammalian
biochronology of the Arikareean through Hemphillian interval (Late Oligocene
through Early Pliocene epochs). Pp. 169–231, In M.O. Woodburn (Ed.). Late Cretaceous
and Cenozoic Mammals of North America. Columbia University Press, New
York, NY. 391 pp.
Vanzolini, P.E. 1952. Fossil snakes and lizards from the lower Miocene of Florida. Journal
of Paleontology 26:452–457.