First definitive record of a stygobiotic fish (Percopsiformes, Amblyopsidae, Typhlichthys) from the Appalachians karst region in the eastern United States

Abstract

In the central and eastern United States, cavefishes have been known historically only from the Interior Low Plateau and Ozarks karst regions. Previously, cavefishes were unknown from the Appalachians karst region, which extends from southeastern New York southwestward into eastern Tennessee, northwestern Georgia, and northeastern Alabama. Here we report the discovery of a new population of the amblyopsid cavefish Typhlichthys subterraneus Girard, 1859 from a cave in Catoosa County, Georgia, that significantly extends the known distribution of the species. The cave is located in the Appalachian Valley and Ridge physiographic province and Appalachians karst region, and represents the first definitive report of a stygobiotic fish from the Appalachians karst region. Genetic analyses of one mitochondrial and one nuclear locus from the cavefish indicate this population is closely allied with populations that occur along the western margins of Lookout and Fox mountains in Dade County, Georgia, and populations to the northwest in southern Marion County, Tennessee. It is likely that these populations are also related to those from Wills Valley, DeKalb County, Alabama. The distribution of this new population of T. subterraneus and its close allies pre-dates the emergence of a Tennessee-Coosa River drainage divide in the Pliocene. The potential exists to discover additional populations in caves within the Appalachians karst region in Catoosa County and northward into Hamilton County, Tennessee.

Full text

First denitive record of a stygobiotic sh (Percopsiformes, Amblyopsidae, Typhlichthys)... 39
First definitive record of a stygobiotic fish
(Percopsiformes, Amblyopsidae, Typhlichthys) from the
Appalachians karst region in the eastern United States
Matthew L. Niemiller1, Kirk S. Zigler2, Pamela B. Hart3, Bernard R. Kuhajda4,
Jonathan W. Armbruster3, Breanne N. Ayala2, Annette S. Engel5
1 Illinois Natural History Survey, University of Illinois Urbana-Champaign, Champaign, IL 61820 2 Department
of Biology, University of the South, Sewanee, TN 37383 3 Department of Biological Sciences, Auburn University,
Auburn, AL 36849 4 Tennessee Aquarium Conservation Institute, Chattanooga, TN 37402 5 Department of
Earth and Planetary Sciences, University of Tennessee, Knoxville, TN 37996
Corresponding author: Matthew L. Niemiller (cavemander17@gmail.com)
Academic editor: O. Moldovan|Received 29 June 2016|Accepted 14 September 2016|Published 8 November2016
http://zoobank.org/5A60148E-4803-42AB-B85F-4F05581D3643
Citation: Niemiller ML, Zigler KS, Hart PB, Kuhajda BR, Armbruster JW, Ayala BN, Engel AS (2016) First denitive
record of a stygobiotic sh (Percopsiformes, Amblyopsidae, Typhlichthys) from the Appalachians karst region in the eastern
United States. Subterranean Biology 20: 39–50. doi: 10.3897/subtbiol.20.9693
Abstract
In the central and eastern United States, caveshes have been known historically only from the Interior
Low Plateau and Ozarks karst regions. Previously, caveshes were unknown from the Appalachians karst
region, which extends from southeastern New York southwestward into eastern Tennessee, northwestern
Georgia, and northeastern Alabama. Here we report the discovery of a new population of the amblyopsid
cavesh Typhlichthys subterraneus Girard, 1859 from a cave in Catoosa County, Georgia, that signicantly
extends the known distribution of the species. e cave is located in the Appalachian Valley and Ridge
physiographic province and Appalachians karst region, and represents the rst denitive report of a sty-
gobiotic sh from the Appalachians karst region. Genetic analyses of one mitochondrial and one nuclear
locus from the cavesh indicate this population is closely allied with populations that occur along the
western margins of Lookout and Fox mountains in Dade County, Georgia, and populations to the north-
west in southern Marion County, Tennessee. It is likely that these populations are also related to those
from Wills Valley, DeKalb County, Alabama. e distribution of this new population of T. subterraneus
and its close allies pre-dates the emergence of a Tennessee-Coosa River drainage divide in the Pliocene.
e potential exists to discover additional populations in caves within the Appalachians karst region in
Catoosa County and northward into Hamilton County, Tennessee.
Subterranean Biology 20: 39–50 (2016)
doi: 10.3897/subtbiol.20.9693
http://subtbiol.pensoft.net
Copyright Matthew L. Niemiller et al. This is an open access article distributed under the terms of the Creative Commons Attribution License
(CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
RESEARCH ARTICLE
Subterranean
Biology Published by
The International Society
for Subterranean Biology
A peer-reviewed open-access journal

Matthew L. Niemiller et al. / Subterranean Biology 20: 39–50 (2016)
40
Keywords
Appalachian Valley and Ridge, Catoosa County, cavesh, Cumberland Plateau, Georgia, range extension
Introduction
Of the more than 50,000 caves reported in the United States, about 30% occur in the
states of Tennessee, Alabama, and Georgia (TAG). e two most biodiverse karst re-
gions in the United States – the Interior Low Plateau (ILP) and Appalachians – occur
in this region (Culver et al. 2000, Culver and Pipan 2009). e ILP is comprised of
horizontal strata of Ordovician through Mississippian age that extend from southern
Illinois and Indiana, southward through Tennessee and Kentucky and into north-
ern Alabama. e escarpments of the Cumberland Plateau in Kentucky, Tennessee,
Alabama, and Georgia are included in the ILP karst region (Culver et al. 2000). e
ILP and Appalachians karst regions are proximal to each other near the junction of
TAG state borders, although the boundary between the ILP region and the Appala-
chians karst region, and Appalachian Valley and Ridge (AVR) physiographic province,
is somewhat arbitrary. Caves in the Appalachians karst are predominantly developed
within Paleozoic rocks of an ancient fold-and-thrust belt associated with compression
during Alleghenian orogenesis of the Appalachian Mountains (Hatcher et al. 2007,
Hatcher 2010). e AVR physiographic province is comprised of parallel ridges of
sandstones with intervening structural valleys of folded and faulted shales and car-
bonates that extend from southeastern New York to eastern Tennessee, northwestern
Georgia, and northeastern Alabama between the Blue Ridge Mountains to the east and
the Appalachian Plateau (specically, the Cumberland Plateau) to the west.
e ILP and Appalachian karst regions contain the most caves and have the great-
est richness of troglobiotic taxa in the United States (Culver et al. 2003, Hobbs 2012).
In particular, a hotspot of subterranean biodiversity and endemism has been identied
near the contact of the ILP and Appalachians karst regions along the escarpments of the
Cumberland Plateau in northeastern Alabama and south-central Tennessee (Culver et al.
1999, 2000, 2006, Christman et al. 2005, Niemiller and Zigler 2013). Species richness in
the Appalachians karst region (and AVR) is less than half that observed in the ILP in the
TAG region, and AVR subterranean fauna are distinct from ILP fauna. Only 9% of the
200+ troglobionts in Tennessee occur in both karst regions (Niemiller and Zigler 2013).
Several factors may explain dierences in species richness between these the ILP
and Appalachians karst regions, such as dierences in habitat availability, habitat con-
nectivity, historical factors, and surface productivity (Christman and Culver 2001,
Culver et al. 2006, Niemiller and Zigler 2013). Cave density has been viewed as a
surrogate for habitat availability and connectivity because it positively correlates with
regional species richness (Christman and Culver 2001, Culver et al. 2003, 2006). Cave
density is considerably lower in the southern Appalachians karst region compared to
the ILP in the TAG region. Moreover, the folded and faulted cave-bearing strata in
the Appalachians karst region are dissected and discontinuous compared to horizontal

First denitive record of a stygobiotic sh (Percopsiformes, Amblyopsidae, Typhlichthys)... 41
Figure 1. Typhlichthys subterraenus collected 25 November 2015 from Crane Cave (GCZ80), Catoosa
County, Georgia. Photograph by B.R. Kuhajda.
strata of the ILP. A major zone of faulting along the eastern escarpment of the Cum-
berland Plateau in the Appalachians karst region has been hypothesized to act as a
stratigraphic barrier to subterranean dispersal between the two karst regions (Barr and
Holsinger 1985, Miller and Niemiller 2008, Niemiller et al. 2008, 2009), which may
explain why so few species occur in both karst regions.
Typhlichthys subterraneus s.l. Girard, 1859 is one of the most wide-ranging cave-
shes in the world (Proudlove 2006, Niemiller and Poulson 2010). In the TAG region,
this cavesh is known from >180 caves in the ILP, with the greatest concentration of
occurrences in central Tennessee and northern Alabama (Niemiller et al. 2013b,c).
In Georgia, T. subterraneus is known only from four caves developed in Mississippian
Bangor Limestone along the western margins of Lookout Mountain and Fox Moun-
tain in Dade County, Georgia (Cooper and Iles 1971, Freeman and Niemiller 2009,
Niemiller et al. 2012a, 2013b,c). Here we report the discovery of a population of the
Southern Cavesh (Typhlichthys subterraneus) from Crane Cave in Catoosa County,
northwestern Georgia (Fig. 1), located in the center of the AVR physiographic prov-
ince and the Appalachians karst region. Not only does this record represent a signi-
cant range extension for this species, but it also represents the rst denitive report of
a stygobiotic sh from the Appalachians karst region.
Materials and methods
Study site
Crane Cave (Georgia Speleological Survey cave no. GCZ80) is located ca. 7 km SSE
of Fort Oglethorpe, Georgia, in the South Chickamauga Creek watershed. Crane Cave
formed in the Ordovician Newala Limestone, and has 292 m of mapped length with
11 m of vertical extent and three entrances. A small stream runs through the cave and
emerges at the spring entrance. e stream begins in a large pool at the back of the

Matthew L. Niemiller et al. / Subterranean Biology 20: 39–50 (2016)
42
cave called “e Found Sea.” e pool is ca.10 m in length and ca. 6 m in width, and
has a mud/silt substrate bottom. e full extent of the pool is unknown, as it extends
underneath a ledge at the back of the cave. At base level, water depth is ca. 2 m deep
in the deepest portion of the pool.
Cavesh survey
Crane Cave was visited on four occasions: 10 August 2015, 18 August 2015, 29 Oc-
tober 2015, and 25 November 2015. e Found Sea and other aquatic habitats were
sampled using time-constrained visual surveys with headlamps and handheld dive
lights. Richness and abundance data for aquatic fauna were recorded, and a concerted
eort was made to capture sh with handheld dipnets. A voucher specimen and tissue
sample (n clip) was obtained for morphological and genetic analyses.
Molecular methods and analyses
Genomic DNA was extracted from n clips using the EZNA DNA Extraction Kit
(Omega Biotek). Two gene loci were chosen from six previously used by Niemiller
et al. (2012b) to determine the genetic identify and relationships of the Crane Cave
population to other Typhlichthys populations. e protein-coding mitochondrial
NADH dehydrogenase 2 (ND2) gene was amplied by PCR with primers TyCon1F
(5’-TGAACCCTTTCATCCTAATAGCC-3’) and TyCon1R (5’-GGTTGTGAG-
GAGGGTCAGG-3’). Each PCR reaction contained 8.5 µL of puried water, 12.5
µL Master Mix (Promega Corporation), 2.0 µL DNA template, 1.0 µL each of 10 µM
forward and reverse primers. Amplication began with an initial denaturation of 94
°C for 30 seconds, followed by 30 cycles of 94 °C denaturing for 30 seconds, annealing
at 51.2 °C for 30 seconds, elongation at 72 °C for 75 seconds, then a nal elongation
step of 10 minutes. e gene sequence was 957 base pairs (bp) long. A 774-bp sec-
tion of the rst intron of the ribosomal nuclear encoded S7 gene was amplied with
the primers S7Con1F (5’-TCTGCAGGATGGAAGATTTTGT-3’) and S7Con1R
(5’-GCTTGTACTGAACATGGCCC-3’). e PCR reactions contained the same
amount and concentration of reagents as the ND2 reaction. e initial denaturation
for amplication began at 95 °C for 60 seconds, followed by 30 cycles of denatura-
tion at 95 °C for 30 seconds, annealing at 60 for 60 seconds, elongation at 72 °C for
2 minutes, followed by nal elongation at 72 °C for 10 minutes. PCR products were
cleaned using ExoSAP-IT (Aymetrix) and bidirectionally sequenced at Genewiz, Inc.
(Cambridge, Massachusetts, USA). Unique sequences generated for the Crane Cave
sample were accessioned into GenBank (ND2: KX173801 and S7: KX246929).
Forward and reverse sequences were aligned into contigs and edited with manu-
al verication using Geneious v. 6.0.6 (Biomatters Ltd.). Maximum likelihood gene

First denitive record of a stygobiotic sh (Percopsiformes, Amblyopsidae, Typhlichthys)... 43
trees were generated for both ND2 and S7 loci with raxmlGUI v.1.31 (Silvestro
and Michalak 2012). Codon partitioning according to Niemiller et al. (2012b)
was utilized for ND2. For both loci, a maximum likelihood + thorough bootstrap
analysis was conducted with 10 replicates of 100 runs utilizing the caveshes Speo-
platyrhinus poulsoni Cooper & Kuehne, 1974 and Amblyopsis spelaea DeKay, 1842
as outgroup taxa.
Results
A single cavesh was observed in e Found Sea of Crane Cave but evaded capture
during an initial bioinventory on 10 August 2015. No cavesh were observed dur-
ing two subsequent trips on 18 August 2015 and 29 October 2015. Two caveshes
were observed on 25 November 2015. One specimen was collected and retained as a
voucher specimen (Fig. 1). e specimen was identied as Typhlichthys subterraneus by
the lack of external eyes (vs. presence in Chologaster and Forbesichthys), presence of one
row of exposed neuromasts on each half of the caudal n (Amblyopsis, Speoplatyrhinus,
and Troglichthys have four to six rows, two to three on each half of the caudal n), the
presence of branched rays in the pectoral ns (vs. unbranched in Speoplatyrhinus), the
lack of pelvic ns (vs. presence in Amblyopsis), and nine dorsal-n rays (vs. 7–8 in Tro -
glichthys). In addition, only Typhlichthys, among stygobiotic amblyopisids, is known to
have an extensive pigment response when exposed to light (Eigenmann 1909; Poulson
1963). e Crane Cave specimen has extensive melanophore development particularly
along the edges of myomeres, on the head, and at the bases of the median ns (Ambly-
opsis, Speoplatyrhinus, and Troglichthys have far fewer melanophores with less melanin,
and color is not generally noticeable in preserved specimens). e specimen was cata-
loged into the Auburn University Museum of Natural History (AUM 67212) and a
tissue sample (n clip) was accessed into the Auburn University Fish Tissue Collection
(AUFT 2651).
Other notable fauna observed during the four biological surveys at Crane Cave
included aquatic species Crangonyx antennatus Cope & Packard, 1881 (Amphipoda:
Crangonyctidae), Caecidotea richardsonae Hay, 1901 (Isopoda: Asellidae), and Cot-
tus sp. (Scorpaeniformes: Cottidae), and terrestrial species Hesperochernes mirabilis
(Banks, 1895) (Pseudoscorpiones: Chernetidae), Bishopella sp. (Opiliones: Phalango-
didae), Amoebaleria sp. (Diptera: Heleomyzidae), and Eidmanella pallida (Emerton,
1875) (Araneae: Nesticidae).
Molecular results indicated that the Crane Cave specimen was most closely re-
lated to the T. subterraneus populations designated lineage A in both the ND2 and
S7 phylogenies (Niemiller et al. 2012b). In the ND2 phylogeny (Fig. 2), the Crane
Cave specimen was sister to a clade containing populations from Long’s Rock Wall
(GDD101) and Limestone Caverns (GDD140) from Dade County, Georgia, in the
Lookout Creek watershed, and the closest populations in geographical proximity to

Matthew L. Niemiller et al. / Subterranean Biology 20: 39–50 (2016)
44
Figure 2. Maximum likelihood gene trees for mitochondrial ND2 (left) and nuclear S7 (right) loci. Colors
correspond to genetic lineages for Typhlichthys subterraenus designated in Niemiller et al. (2012b). Boot-
strap values are to the left (ND2) or right (S7) of the corresponding node with >0.70 support. Outgroup
taxa include Speoplatyrhinus poulsoni and Amblyopsis hoosieri. Scale bar unit: expected substitutions per site.
Crane Cave (Fig. 3). e clade comprised of Crane Cave, Long’s Rock Wall, and
Limestone Caverns was sister to a population from Pryor Cave Spring (Tennessee
Cave Survey no. TMN129) and Lost Pig Cave (TMN20) located in the Little Se-
quatchie River Valley and Sweetens Cove of southern Marion County, Tennessee,
respectively. In contrast, the base of lineage A in the S7 phylogeny was a strongly
supported polytomy that consisted of Crane Cave, Pryor Cave Spring, and a Long’s
Rock Wall + Limestone Caverns clade (Fig. 2). e ND2 and the S7 phylogenies both
presented strong support for the monophyly of lineage A.

First denitive record of a stygobiotic sh (Percopsiformes, Amblyopsidae, Typhlichthys)... 45
Figure 3. Distribution of Typhlichthys subterraneus (solid circles) in southeastern Tennessee, northeastern
Alabama, and northwestern Georgia. e new record at Crane Cave is denoted with a red triangle and lin-
eage A localities are highlighted in peach. Lineage A populations that have been genetically examined are
marked with an asterisk and labeled as follows: LMC – Limestone Caverns, LPC – Lost Pig Cave, LRW
– Long’s Rock Wall, and PCS – Pryor Cave Spring. Counties with Typhlichthys records are labeled. Karst
and cave-bearing strata are shaded gray based on the U.S. karst map (Weary and Doctor 2014). e border
of the Appalachian Valley and Ridge (AVR) physiographic province is denoted by the dot and dashed line.
Discussion
e range of Typhlichthys subterraneus s.l. extends throughout the ILP of Kentucky,
Tennessee, Alabama, and Georgia, which makes it one of the largest distributions of
any cavesh in the world (Proudlove 2006, Niemiller and Poulson 2010). Because of
the widespread distribution, even from distinct hydrological basins, several authors hy-
pothesize that T. subterraenus represents a complex of morphologically cryptic, but ge-
netically distinct, species (Swoord 1982; Barr and Holsinger 1985; Holsinger 2000;
Niemiller and Fitzpatrick 2008; Niemiller and Poulson 2010). Niemiller et al. (2012b)
identify at least ten cryptic lineages from a species delimitation analysis based on six
loci and samples from 60 populations across the range. e most recent common an-
cestor of these lineages dates to the Late Pliocene to Early Pleistocene, about 2.8 mil-
lion years ago (Mya) (95% condence interval: 2.1–3.5 Mya; Niemiller et al. 2013a).
Populations from Dade County, Georgia (Limestone Caverns and Long’s Rock Wall),
and at least two populations from the Little Sequatchie River Valley in Tennessee, form

Matthew L. Niemiller et al. / Subterranean Biology 20: 39–50 (2016)
46
a distinct genetic Typhlichthys lineage, referred to as lineage A. Populations that occur
in Wills Valley in DeKalb County, Alabama, also are thought to belong to lineage A
(Niemiller et al. 2013b), but have not been genetically examined to date. is lineage
diverged from others in the ILP about 2.2 Mya (1.6–2.9 Mya based on 95% con-
dence intervals; Niemiller et al. 2012b, 2013a).
Analyses of the mitochondrial ND2 and the nuclear S7 loci from Crane Cave
T. subterraneus strongly support anity to lineage A (as dened by Niemiller et al.
2012b). However, the new Crane Cave record is ca. 24.2 km straight-line distance to
the east from the next closest populations in Georgia and Alabama. Specically, the T.
subterraneus populations in Dade County are from caves formed in the Mississippian-
age Bangor Limestone on the escarpments of Lookout Mountain and Fox Mountain,
clearly within the Cumberland Plateau physiographic province. Despite the arbitrary
boundary between the ILP and AVR, the distribution of lineage A now extends from
the ILP into the Appalachians karst region because Crane Cave is well within the AVR
and is from a hydrologically distinct watershed compared to the previously described
T. subterraneus populations in the TAG region (Fig. 3).
ere is the issue of whether the other T. subterraneus populations in lineage A, spe-
cically those in Wills Valley formed in Cambrian-Ordovician Knox group dolomites
in AVR-style structural valleys, are also considered AVR distributions or ILP distribu-
tions. e physiographic distinction of Wills Valley has been a matter of debate in the
literature. Wills Valley is an anticlinal valley anked by Sand Mountain to the west and
Lookout Mountain to the east. Both ridges are considered parts of the Cumberland
Plateau (Johnson 1930, Harkins et al. 1982, Raymond et al. 1988). As such, previous
studies comparing subterranean biodiversity among karst regions have considered Wills
Valley to be associated with the Cumberland Plateau and ILP karst region rather than the
AVR within the Appalachians karst region (Peck 1989, 1995, Culver et al. 2003, Hobbs
2012). However, others have placed Wills Valley as part of the Ridge and Valley Level III
ecoregion (Grith et al. 2001) based on ecosystem similarity according to land use, ge-
ology, physiography, hydrology, climate, natural vegetation, and soils (Omernik 1987).
Regardless, the distribution of T. subterraneus in Wills Valley caves and the evolution
of the karst in the valley warrant further study. e transitional location of Wills Valley
between the ILP and Appalachians karst region and its length (100+ km) may have been
critical in the movement of T. subterraneus between the two larger karst regions.
Another important aspect of T. subterraneus in Wills Valley is that these popula-
tions are in the Coosa River watershed, which ows into the Alabama River and then
Mobile Bay. Crane Cave occurs in the South Chickamauga Creek watershed, which
ows into the Tennessee River. Moreover, all four documented populations in Dade
County, Georgia, occur in the Lookout Creek watershed, and the caves in Marion
County, Tennessee, are part of the Sequatchie River watershed. Both Lookout Creek
and the Sequatchie River empty into the Tennessee River, which eventually ows
into the Ohio River and then the Mississippi River. River drainages in the southern
region of North America and the Appalachian Mountains became established at least
by the Eocene, 55 Mya (Galloway et al. 2011, Hoagstrom et al. 2013). At this time,
the ancestral Tennessee River and the Coosa River formed the Appalachian River that

First denitive record of a stygobiotic sh (Percopsiformes, Amblyopsidae, Typhlichthys)... 47
owed to Mobile Bay (Johnson 1905, Milici 1968). By the mid-Miocene through the
Pliocene, uplift in the Southern Appalachians (Gallen et al. 2013) or of the Nashville
Dome (Clark 1989), as well as potential regional base-level lowering, initiated down-
cutting by the ancestral Tennessee River through Walden Ridge and westward ow
into the Sequatchie Valley, then around the Nashville Dome before being captured
by the Ohio River (Milici 1968, Clark 1989, Self 2000). Some suggest that stream
capture may have been facilitated by karst as “cavern capture” (s.s. Johnson 1905) in
Walden Ridge and the Sequatchie Valley.
Today, the Tennessee and Coosa rivers are separated by a divide, whereby the south-
ern part of Wills Valley ows to the Coosa River and the northern section ows to the
Tennessee River. e genetic aliation of the Crane Cave T. subterraneus population to
lineage A (Niemiller et al. 2012b) suggests that this lineage has a shared evolutionary his-
tory, whereby a common ancestor must pre-date the emergence of the Tennessee-Coosa
drainage divide and subsequent isolation of the Tennessee River from the Coosa River.
e drainage divide likely formed in the late Pliocene based on evidence from changes in
deltaic sedimentation (Galloway et al. 2011) and age dates from cave sediment records
(Anthony and Granger 2007). is timeframe corresponds with the estimated diver-
gence of lineage A from 2.9 to 1.6 Mya from other lineages in the ILP (Niemiller et al.
2012b, 2013a). Continued uplift and stream incision further isolated lineage A popula-
tions throughout the TAG region in the early Pleistocene, which resulted in genetically
distinct populations in Crane Cave, Dade County, and Alabama/Tennessee.
In conclusion, although no additional cavesh populations have been discovered
in the past several years (Freeman and Niemiller 2009), with the exception of Crane
Cave, and despite several cave bioinventories and other studies in northwestern Geor-
gia (Reeves et al. 2000, Buhlmann 2001, Miller and Niemiller 2008), the potential
exists for additional T. subterraneus populations to be discovered. Caves to be targeted
for exploration would be those within the South Chickamauga Creek watershed and
formed in the Newala Limestone throughout Catoosa County, as well as extending
southward into Walker County and northward into Hamilton County, Tennessee.
Lastly, these T. subterraneus populations may provide insight into the geologic history
of the Tennessee and Coosa rivers, as well as aid in the understanding of other endemic
cave fauna in the TAG region. In particular, the boundary between the ILP and Ap-
palachians karst regions (and Interior Plateau and AVR physiographic provinces) may
not be as strong a barrier to dispersal for stygobiotic taxa as previously thought.
Acknowledgements
We particularly thank Mary and Ron Ziegler for allowing us to visit Crane Cave and for
welcoming us on several return trips. Funding and support for this project was provided
by the Cave Conservancy Foundation and the University of the South. Wethank the
Georgia Speleological Survey for providing data and a map of Crane Cave. is work was
permitted by the Georgia Department of Natural Resources under scientic collection
permit no. 8934 and approval by the University of the South IACUC committee (KSZ).

Matthew L. Niemiller et al. / Subterranean Biology 20: 39–50 (2016)
48
References
Barr TC, Holsinger JR (1985) Speciation in cave faunas. Annual Review of Ecology, Evolution,
and Systematics 16: 313–337. doi: 10.1146/annurev.es.16.110185.001525
Buhlmann KA (2001) A biological inventory of eight caves in northwestern Georgia with con-
servation implications. Journal of Cave and Karst Studies 63: 91–98.
Christman MC, Culver DC (2001) e relationship between cave biodiversity and available
habitat. Journal of Biogeography 28: 367–380. doi: 10.1046/j.1365-2699.2001.00549.x
Christman MC, Culver DC, Madden M, White D (2005) Patterns of endemism of the eastern
North American cave fauna. Journal of Biogeography 32: 1441–1452. doi: 10.1111/j.1365-
2699.2005.01263.x
Clark GM (1989) Central and Southern Appalachian water and wind gap origins: Review and
new data. Geomorphology 2: 209–232. doi: 10.1016/0169-555X(89)90013-5
Cooper JE, Iles A (1971) e Southern Cavesh, Typhlichthys subterraneus, at the southeastern
periphery of its range. National Speleological Society Bulletin 33: 45–49.
Cressler CW (1963) Geology and ground-water resources of Catoosa County, Georgia. Infor-
mation Circular no. 28. e Geological Survey, Georgia State Division of Conservation,
Atlanta, Georgia, 19 pp.
Cressler CW (1964) Geology and ground-water resources of Walker County, Georgia. Infor-
mation Circular no. 29. e Geological Survey, Georgia State Division of Conservation,
Atlanta, Georgia. 15 pp.
Croft MG (1964) Geology and ground-water resources of Dade County, Georgia. Information
Circular no. 29. e Geological Survey, Georgia State Division of Conservation, Atlanta,
Georgia, 17 pp.
Culver DC, Christman MC, Elliott WR, Hobbs HH III, Reddell JR (2003) e North Ameri-
can obligate cave fauna: regional patterns. Biodiversity and Conservation 12: 441–468.
doi: 10.1023/A:1022425908017
Culver DC, Deharveng L, Bedos A, Lewis JJ, Madden M, Reddell JR, Sket B, Trontelj P,
White D (2006) e midlatitude biodiversity ridge in terrestrial cave fauna. Ecography 29:
120–128. doi: 10.1111/j.2005.0906-7590.04435.x
Culver DC, Hobbs HH III, Mylroie JE (1999) Alabama: a subterranean biodiversity hotspot.
Journal of the Alabama Academy of Science 70: 97–104.
Culver DC, Master LL, Christman MC, Hobbs HH III (2000) Obligate cave fauna of the
48 contiguous United States. Conservation Biology 14: 386–401. doi: 10.1046/j.1523-
1739.2000.99026.x
Culver DC, Pipan T (2009) e biology of caves and other subterranean habitats. Oxford
University Press, Oxford, 254 pp.
Eigenmann CH (1909) Cave Vertebrates of America. A Study in Degenerative Evolution.
Carnegie Institution of Washington, Washington, 241 pp.
Freeman BJ, Niemiller ML (2009) Species prole for Southern Cavesh, Typhlichthys subter-
raneus. Georgia Rare Species Proles. Georgia Department of Natural Resources, Wildlife
Resources Division, Atlanta, Georgia.
Gallen SF, Wegmann KW, Bohnenstiehl DR (2013) Miocene rejuvenation of topographic
relief in the southern Appalachians GSA Today 23: 4–10. doi: 10.1130/GSATG163A.1

First denitive record of a stygobiotic sh (Percopsiformes, Amblyopsidae, Typhlichthys)... 49
Galloway WE, Whiteaker TL, Ganey-Curry P (2011) History of Cenozoic North American
drainage basin evolution, sediment yield, and accumulation in the Gulf of Mexico basin.
Geosphere 7: 938–973. doi: 10.1130/GES00647.1
Grith GE, Omernik JM, Comstock JA, Lawrence S, Martin G, Goddard A, Hulcher, VJ,
Foster T (2001) Ecoregions of Alabama and Georgia, (color poster with map, descriptive
text, summary tables, and photographs). U.S. Geological Survey, Reston, Virginia.
Harkins JR (1982) Hydrology of Area 24, Eastern Coal Province, Alabama and Georgia. US
Geological Survey Open-File Report 81-1113, 81 pp.
Hatcher RD Jr, Bream BR, Merschat AJ (2007) Tectonic map of the southern and central
Appalachians: A tale of three orogens and a complete Wilson cycle. GSA Memoirs 200:
595–632. doi: 10.1130/2007.1200(29)
Hatcher RD Jr (2010) e Appalachian orogen: A brief summary. GSA Memoirs 206: 1–19.
doi: 10.1130/2010.1206(01)
Hoagstrom CW, Ung V, Taylor K (2013) Miocene rivers and taxon cycles clarify the com-
parative biogeography of North American highland shes. Journal of Biogeography 41:
644–658. doi: 10.1111/jbi.12244
Hobbs HH III (2012) Diversity patterns in the United States. In: White WB, Culver DC (Eds)
Encyclopedia of caves (2nd edn). Academic Press, Amsterdam, 251–264. doi: 10.1016/
b978-0-12-383832-2.00033-5
Holsinger JR (2000) Ecological derivation, colonization, and speciation. In: Wilkens H, Cul-
ver DC, Humphreys WF (Eds) Ecosystems of the World (Vol 30), Subterranean ecosys-
tems. Elsevier, Amsterdam, 399–415.
Johnson DW (1905) e Tertiary history of the Tennessee River. Journal of Geology 13:
194–231. doi: 10.1086/621220
Johnson WD (1930) Physical divisions of northern Alabama (Bulletin 38). Geological Survey
of Alabama, Tuscaloosa, Alabama, 48 pp.
Milici R (1968) Mesozoic and Cenozoic physiographic development of the Lower Tennessee
River: In terms of the dynamic equilibrium concept. Journal of Geology 76: 472–479. doi:
10.1086/627345
Miller BT, Niemiller ML (2008) Distribution and relative abundance of Tennessee Cave Sala-
manders (Gyrinophilus palleucus and G. gulolineatus) with an emphasis on Tennessee popu-
lations. Herpetological Conservation and Biology 3: 1–20.
Niemiller ML, Fitzpatrick BM (2008) Phylogenetics of the Southern Cavesh (Typhlichthys
subterraneus): implications for conservation and management. Proceedings of the 18th Na-
tional Cave and Karst Management Symposium, St. Louis, Missouri, 79–88.
Niemiller ML, Poulson TL (2010) Subterranean shes of North America: Amblyopsidae. In:
Trajano E, Bichuette ME, Kappor BG (Eds) e biology of subterranean shes. Science
Publishers, Eneld, 169–280. doi: 10.1201/EBK1578086702-c7
Niemiller ML, Zigler KS (2013) Patterns of cave biodiversity and endemism in the Appalachians
and Interior Plateau of Tennessee USA. PLoS ONE 8: e64177. doi: 10.1371/journal.
pone.0064177
Niemiller ML, Fitzpatrick BM, Miller BT (2008) Recent divergence with gene ow in Ten-
nessee Cave Salamanders (Plethodontidae: Gyrinophilus) inferred from gene genealogies.
Molecular Ecology 17: 2258–2275. doi: 10.1111/j.1365-294X.2008.03750.x

Matthew L. Niemiller et al. / Subterranean Biology 20: 39–50 (2016)
50
Niemiller ML, Miller BT, Fitzpatrick BM (2009) Systematics and evolutionary history of sub-
terranean Gyrinophilus salamanders. Proceedings of the 15th International Congress of Spe-
leology, Kerrville, Texas 15: 242–248.
Niemiller ML, Fenolio DB, Zigler KS (2012a) e obligate cave fauna of Georgia. Bulletin of
the Georgia Speleological Survey 2012: 6–12.
Niemiller ML, Near TJ, Fitzpatrick BM (2012b) Delimiting species using multilocus data:
diagnosing cryptic diversity in the Southern Cavesh Typhlichthys subterraneus (Teleostei:
Amblyopsidae). Evolution 66: 846–866. doi: 10.1111/j.1558-5646.2011.01480.x
Niemiller ML, Fitzpatrick BM, Shah P, Schmitz L, Near TJ (2013a) Evidence for repeated loss
of selective constraint in rhodopsin of amblyopsid caveshes (Teleostei: Amblyopsidae).
Evolution 67: 732–748. doi: 10.1111/j.1558-5646.2012.01822.x
Niemiller ML, Graening GO, Fenolio DB, Godwin JC, Cooley JR, Pearson WR, Near TJ,
Fitzpatrick BM (2013b) Doomed before they are described? e need for conservation
assessments of cryptic species complexes using an amblyopsid cavesh (Amblyopsidae: Ty-
phlichthys) as a case study. Biodiversity and Conservation 22: 1799–1820. doi: 10.1007/
s10531-013-0514-4
Niemiller ML, Zigler KS, Fenolio DB (2013c) Cave life of TAG: a guide to commonly en-
countered species in Tennessee, Alabama, and Georgia. Biology Section of the National
Speleological Society, Huntsville, Alabama, 45 pp.
Omernik JM (1987) Ecoregions of the conterminous United States. Annals of the Association
of American Geographers 77: 118–125. doi: 10.1111/j.1467-8306.1987.tb00149.x
Peck SB (1989) e cave fauna of Alabama: part I. e terrestrial invertebrates (excluding
insects). National Speleological Society Bulletin 51: 11–33.
Peck SB (1995) e cave fauna of Alabama: part II – e insects. National Speleological Soci-
ety Bulletin 57: 1–19.
Poulson TL (1963) Cave adaptation in amblyopsid shes. American Midland Naturalist 70:
257–290. doi: 10.2307/2423056
Proudlove GS (2006) Subterranean shes of the world. International Society for Subterranean
Biology, Moulis, 300 pp.
Raymond DE, Osborne WE, Copeland CW, Neathery TL (1988) Alabama stratigraphy. Cir-
cular 140. Geological Survey of Alabama, Stratigraphy and Paleontology Division, Tus-
caloosa, 97 pp.
Reeves WK, Jensen JB, Ozier JC (2000) New faunal and fungal records from caves in Georgia,
USA. Journal of Cave and Karst Studies 62: 169–179.
Self RP (2000) e pre-Pliocene course of the lower Tennessee River as deduced from river
terrace gravels in southwest Tennessee. Southeastern Geology 39: 61–70.
Silvestro D, Michalak I (2012) raxmlGUI: a graphical front-end for RAxML. Organisms Di-
versity & Evolution 12: 335–337. doi: 10.1007/s13127-011-0056-0
Swoord DL (1982) Genetic variability, population dierentiation, and biochemical relation-
ships in the family Amblyopsidae. Master’s esis, Eastern Kentucky University, Rich-
mond, Kentucky.
Weary DJ, Doctor DH (2014) Karst in the United States: a digital map compilation and data-
base. U.S. Geological Survey Open-File Report 2014-1156, 23 pp.