Evidence of Stream Capture from the Tallapoosa River
Drainage by a Chattahoochee River Tributary Based on Fish
Distributions
Andrew Jarrett, Warren Stiles, Alexis Janosik, Rebecca Blanton, and Carol Johnston
Southeastern Naturalist, Volume 16, Issue 1 (2017): 117–136
Full-text pdf (Accessible only to subscribers.To subscribe click here.)
Southeastern Naturalist
117
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
22001177 SOUTHEASTERN NATURALIST 1V6o(1l.) :1161,7 N–1o2. 61
Evidence of Stream Capture from the Tallapoosa River
Drainage by a Chattahoochee River Tributary Based on Fish
Distributions
Andrew Jarrett1, Warren Stiles1, Alexis Janosik2, Rebecca Blanton3, and
Carol Johnston1,*
Abstract - Based on the distribution of 2 fish species and geological ev idence, we propose
stream capture of a Tallapoosa River tributary by Wehadkee Creek, a tributary of the Chattahoochee
River in east-central Alabama. Micropterus tallapoosae (Tallapoosa Bass) and
Cyprinella gibbsi (Tallapoosa Shiner), endemics to the Tallapoosa River drainage, are found
in Wehadkee Creek (Chattahoochee River drainage). We used mitochrondrial DNA to compare
the Wehadkee Creek specimens of Tallapoosa Shiner to those analyzed in a previous
study of the genetic structure of the species throughout the Tallapoosa River drainage. Their
identity as Tallapoosa Shiner was validated, and we found some divergence relative to other
populations in the Wehadkee Creek fish. We validated the identity of Tallapoosa Bass and
Micropterus chattahoochae (Chattahoochee Bass), using mitochondrial DNA sequences
subjected to phylogenetic analyses of all Micropterus coosae (Redeye Bass) group species
previously identified. In addition to these fish distributions, the geology of the upper Wehadkee
Creek area suggests a past stream capture may have occurred. Alternatively, these
fishes could have been introduced into adjoining drainages by hu mans.
Introduction
Distributions of freshwater organisms can offer clues to past drainage configuration
(Mayden 1988). In some cases, a species with a broad distribution in one drainage
can be found in a small portion of an adjacent drainage. If the pattern of distribution is
repeated in other species, a case can be made for a chance event, such as stream capture,
playing a role in the distribution of these species (Wiley 1988).
The process of streams being displaced from one drainage to another is
termed river capture (Bishop 1995). The distribution and relationships among
fishes have been used to infer stream capture and drainage modification in several
river systems. One of the most obvious indications of geologic drainage
modification based on distribution of fishes are the 11 species of Mobile Basin
endemic fishes found in Bear Creek, Alabama and Mississippi, which currently
flows into the Tennessee River (Wall 1968). These fishes are found nowhere else
in the Tennessee River drainage, strongly suggesting stream capture of a Mobile
Basin stream by a Tennessee River tributary. However, not all distributional
1Fish Biodiversity Lab, School of Fisheries, Aquaculture and Aquatic Sciences, Auburn
University, Auburn, AL 36849. 2University of West Florida, Department of Biology, Building
58/60, 11000 University Parkway, Pensacola, FL 32514. 3Department of Biology, Austin
Peay State University, PO Box 4718, 681 Summer Street, Clarksville, TN 37040. *Corresponding
author - Johnsc5@auburn.edu.
Manuscript Editor: Benjamin Keck
Southeastern Naturalist
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
118
evidence of drainage modification is as striking, and other lines of evidence are
brought to bear.
Most modern studies testing hypotheses regarding past river capture use genetic
analysis of fishes or other aquatic organisms to investigate phylogeographic
relationships within a single species (Howard and Morgan 1993, Hurwood and
Hughes 1998, Waters et al. 2001). Waters et al. (2001) used mitochondrial DNA
sequence data to examine a geomorphic hypothesis used to explain the distribution
of a species complex of freshwater Galaxias. The data suggested that an isolated
population otherwise restricted to Clutha/Kawarau River system found in the Nevis
River diverged at approximately the same time as a water-flow reversal, resulting
in headwater capture. Some studies have used genetic relationships within multiple
species to test hypotheses regarding river capture (Burridge et al. 2006). Using
mitochondrial DNA of 2 galaxiids, Burridge et al. (2006) provided evidence of
stream capture. Divergence between 2 sister species was low in streams thought to
have undergone stream capture, while in other areas divergence between study species
was high. Typically the scale of these studies includes several streams or river
reaches, although some focus on a single stream (Wall 1968). Combining evidence
from the distribution and evolutionary relationships of fishes has enabled the inference
of several river-modification events throughout the world.
We hypothesize that the distribution of 2 fish species in east-central Alabama
may be the result of past stream capture. We documented 2 species (Micropterus
tallapoosae Baker, Johnston, and Blanton [Tallapoosa Bass] and Cyprinella gibbsi
(Howell and Williams) [Tallapoosa Shiner]) endemic to the Tallapoosa River drainage
in a single stream, Wehadkee Creek, of the Chattahoochee River drainage,
Alabama. The presence of Tallapoosa Shiner in Wehadkee Creek was previously
documented, but thought to be a bait-bucket introduction (Boschung and Mayden
2004). During a survey for Micropterus chattahoochae Baker, Johnston, and Blanton
(Chattahoochee Bass), a Chattahoochee River endemic, in Wehadkee Creek,
we documented the occurrence of Tallapoosa Bass. Given that there are 2 species
involved, we propose that a headwater stream capture may have occurred between
Wehadkee Creek (Chattahoochee River drainage) and a tributary of the Tallapoosa
River drainage. The alternative explanation is that these species were separately
introduced into streams where they were not indigenous. We present data on the
distribution of Tallapoosa Bass, Chattahoochee Bass, and Tallapoosa Shiner in Wehadkee
Creek, AL. Genetic analyses are also presented to confirm the identity of
focal species and examine phylogeographic structure.
Field-Site Description
Wehadkee Creek (Fig. 1) was chosen as a survey site for Chattahoochee Bass to
fill knowledge gaps regarding its distribution in Alabama, where it was not known
to occur. Prior surveys for basses in other major Chattahoochee River tributaries
did not detect species of the Micropterus coosae Hubbs and Bailey (Redeye Bass)
species group (Macenia et al. 2007). A search for museum records of species from
this group revealed 2 records: 1 specimen collected in 2001 and 1 from a 1972
Southeastern Naturalist
119
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
collection. Both were identified as Micropterus coosae because M. tallapoosae
and M. chattahoochee were not recognized at the time. This 3rd-order stream originates
as a spring just north of High Shoals, and flows through piedmont upland
Figure 1. Sampling sites in the High Pine and Wehadkee watersheds.
Southeastern Naturalist
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
120
physiography before entering West Point Reservoir on the Chattahoochee River
in east-central Alabama (Fig. 1). The size and upland physiography of Wehadkee
Creek were assumed to be suitable habitat for Chattahoochee Bass, but a recent
survey of stream fishes was lacking.
High Pine Creek (Tallapoosa River drainage) was sampled for Tallapoosa Shiner
to provide material for comparison to Tallapoosa Shiner from Wehadkee Creek.
Samples from this stream were not included in the analysis by Connelly et al.
(2006). Our sample sites in this upland stream were taken from 2nd-order sites in
close proximity to Wehadkee Creek.
Methods
We sampled 9 sites in Wehadkee Creek (Chattahoochee River drainage) for
Chattahoochee Bass (Fig. 1, Table 1). In addition, we sampled a nearby tributary
of the Tallapoosa River, High Pine Creek (HP1, HP2), for Tallapoosa Shiner for
genetic analysis (n = 2 sites). Basses were collected using a backpack electrofisher,
anesthetized using MS 222, photographed, and measured (standard length, SL,
mm). We used seines to collect Tallapoosa shiners. Prior to anesthetizing and releasing
all target species captured, we biopsied fins and placed the samples in 95%
ethanol for genetic analysis. Specimens were vouchered at the Fish Biodiversity
Lab. Treatment and handling methods were approved by Auburn University animal
protocol #2012-2166.
Using the DNeasy Blood and Tissue Kit (Qiagen, Inc.) and following the
manufacturer’s directions, we extracted genomic DNA from fins of 4 specimens of
Tallapoosa Bass and 2 specimens of Chattahoochee Bass from Wehadkee Creek.
Resulting DNA was sequenced for the mitochondrial NADH subunit 2 gene (ND2)
and compared to individuals sequenced for a prior study of Redeye Bass systematics
(Baker et al. 2013) to validate species identifications. Methods of PCR
amplification and sequencing followed those of Baker et al. (2013). We obtained
comparative sequences, including outrgroup taxa (Micropterus salmoides (Lacepède)
[Largemouth Bass]) generated by Baker et al. (2013), from GenBank.
Table 1. Collection sites. Site numbers correspond to those in F igure 1.
Site Description County State Latitude Longitude Date
W1 Wehadkee Creek at Rock Mills Randolph AL 33.15744 -85.28846 06/08/15
W2 Little Wehadkee Creek at CR 20 Heard GA 33.15056 -85.24068 06/16/15
W3 Wehadkee Creek at CR 30 Randolph AL 33.19784 -85.27825 06/16/15
W4 Wehadkee Creek at CR 638 Randolph AL 33.22586 -85.32030 06/10/15
W5 High Shoals Falls (below) Randolph AL 33.23682 -85.33214 07/06/15
W6 High Shoals Falls (above) Randolph AL 33.23783 -85.33383 07/06/15
W7 Unnamed tributary at CR 645 Randolph AL 33.25327 -85.33121 06/16/15
W8 Wehadkee Creek at CR 634 Randolph AL 33.25343 -85.32271 06/10/15
W9 Wehadkee Creek at CR 633 Randolph AL 33.26336 -85.30775 06/16/15
HP1 High Pine Creek at CR 16 Randolph AL 33.20264 -85.34325 06/10/15
HP2 High Pine Creek at CR 59 Randolph AL 33.23132 -85.35699 07/06/15
Southeastern Naturalist
121
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
We used data from 38 Wehadkee Creek and High Pine Creek Tallapoosa Shiner,
along with data from Connelly et al. (2006), in a genetic analysis. Specimens from
these streams were not included in their study. Again using the DNeasy Blood and
Tissue Kit and following the manufacturer’s directions, we extracted genomic
DNA. The mitochondrial NADH 4L dehydrogenase (ND4L) gene of 312-bp was
amplified using polymerase chain reaction (PCR). We sequenced amplified products
bi-directionally with GenWiz, Inc. (New Brunwick, NJ). We downloaded from
Genbank (http://www.ncbi.nlm.nih.gov/genbank/) sequences (Haplotypes A-L)
from Connelly at al. (2006) of ND4L to include in the analysis, and used Cyprinella
trichroistia (Jordan and Gilbert) (Tricolor Shiner) as an outgroup. See Connelly et
al (2006) for details on PCR and sequencing methodology .
We examined the phylogenetic and phylogeographic structure within the Tallapoosa
Shiner and Redeye Bass species group using Bayesian analyses. Prior to
the Bayesian analysis, we deteremined the optimal model of sequence evolution
for each data set (ND4L and ND2, respectively) by evaluation of likelihood scores
for 56 progressively complex models using MrModeltest v2 (Nylander 2004). The
best-fit model and its parameters (GTR+G for both) selected under the Akaike
Information Criterion (AIC) were implemented in MrBayes 3.1.1 (Ronquist and
Huelsenbeck 2003). We used 4 Markov chains with flat priors in all analyses
and started each chain with random trees. Runs consisted of 10 million generations
of Markov chain Monte Carlo (MCMC) simulations. We conducted 2 replicate
runs to ensure the MCMC went through a sufficient number of iterations to allow
convergence in the estimations of tree topology with the best maximum likelihood
posterior probability. The burn-in of the MCMC analysis was determined by graphically
examining the ML scores at each of the sampled generations to find where
values converged. We discarded all trees recorded prior to the burn-in and used the
remaining trees to compute a majority rule consensus tree. Posterior probabilities
(pp) were used as an indication of node support.
Results
We collected Tallapoosa Bass (n = 10) at 6 sites in Wehadkee Creek, and found
Chattahoochee Bass (n = 2) at 1 site (Fig. 1, Table 1). We genetically analyzed 6
specimens. Tallapoosa Bass were found at our uppermost sample sites (W4 and
W8), while Chattahoochee Bass were found downstream at site W1. Tallapoosa
Shiners were found at most sites downstream of High Shoals Fall s (Fig. 1).
Relationships among redeye basses are discussed in detail in Baker et al. (2013).
The analysis conducted herein recovered the same relationships as observed in
that previous study and confirmed species identifications based on examination
of morphological traits typical of the 2 species collected in Wehadkee Creek, Tallapoosa
Bass and Chattahoochee Bass (Fig. 2). Tallapoosa Bass specimens had
green caudal fins, soft dorsal and anal fins, 11 lateral blotches, and no tooth patch.
Chattahoochee Bass had orange tips on caudal fins, soft dorsal and anal fins, 9 lateral
blotches, and tooth patches of 2.0 mm and 24 mm. All individuals identified
Southeastern Naturalist
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
122
by these morphological traits as either Tallapoosa Bass or Chattahoochee Bass from
Wehadkee Creek were recovered in their respective species clades stemming
from analyses of the mitochondrial ND2 gene (Fig. 2).
Although some divergence is evident from Tallapoosa Shiner specimens
found at Wedhadkee Creek and High Pine Creek, the majority of individuals are
closely related to other populations of Tallapoosa Shiner from Connelly et al.
2006 (Fig. 3). Moreover, genetic divergence is low between most individuals
from Wedhadkee Creek, High Pine Creek, and the Tallapoosa River locations,
resulting in low levels of geographic structure in the gene tree. We included
data from High Pine Creek, which is in very close proximity to Wehadkee Creek
(Fig. 1) to see if there was a genetic signature uniting these 2 populations, which
might suggest an origin of the Tallapoosa Shiner population in Wehadkee Creek.
However, we did not see a closer relationship between these 2 populations and
other sources of Tallapoosa Shiner.
Figure 2. Majority rule consensus resulting from the post-burnin Bayesian analysis of the
ND2 gene for members of the Redeye Bass species group. Letters following species names
correspond to those in Table 1 of Baker et al. (2013). Those generated herein, from Wehadkee
Creek labeled using site numbers (W1, W4, and W8) in bold that correspond to site
information in Table 1 and Figure 1. Asterisks indicate well-supported nodes with posterior
probabilities of 0.98 or higher. Outgroup taxa not shown.
Southeastern Naturalist
123
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
Discussion
The distribution of Tallapoosa Bass and Tallapoosa Shiner in Wehadkee Creek
(Chattahoochee River drainage) suggests a potential stream-capture event involving
the Tallapoosa and Chattahoochee drainages in east-central Alabama, as these
species are otherwise found only in the Tallapoosa River drainage. Although one of
these species is a minnow, and could be the result of a bait-bucket transfer, the other
is a species of non-game bass, a less likely candidate for human introduction.
The occurrence of 2 species in drainages outside their much larger ranges can be
considered a repeated pattern, evidence for a chance event (Wiley 1988). Wall (1968)
provided both geologic and distribution data for 11 taxa of fishes that supported
Figure 3. Bayesian inference topology for ND4L sequence data of Tallapoosa Shiner.
Number next to node indicates posterior probabilities. Site numbers correspond to those in
Table 1 and Figure 1.
Southeastern Naturalist
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
124
headwater capture of the Bear Creek system by the Tennessee River. Similarly, Jenkins
et al. (1972) documented the transfer of 6 species from the Casselman River
drainage to the Potomac River via distributional and geologic evidence. Other studies
of possible stream captures have examined the genetic relationships of single
species to trace potential transfers by looking at patterns of genetic similarity and
divergence (Howard and Morgan 1993, Strange 1998). We used genetic analysis
to verify the identity of Tallapoosa Shiner and Tallapoosa Bass, and to examine
potential patterns of genetic relatedness within Tallapoosa Shiner (where there was
sufficient sample size). Although we found some divergence among individuals of
Tallapoosa Shiner from the Wehadkee Creek system, it is uninformative in regards
to the drainage relationships.
The distribution of a third species might also be explained by the same streamcapture
event. Luxilus zonistius Jordan (Bandfin Shiner) is widely distributed
throughout the Chattahoochee River drainage and is also found in a much smaller
number of streams in the upper Tallapoosa River drainage (Boschung and Mayden
2004). Boschung and Mayden (2004) proposed introduction by humans as an
explanation for the occurrence of Bandfin Shiner in Tallapoosa River streams.
However, we suggest that given the location of Bandfin Shiner populations in relation
to Wehadkee Creek (very close proximity), this species may have gained access
to the Tallapoosa River drainage via the same capture event or similar repeated
capture events that may have transferred Tallapoosa Bass and Tallapoosa Shiner
into the Chattahoochee River drainage.
Further evidence of a potential stream-capture event comes from the geological
record. Upper Wehadkee Creek (Chattahoochee River drainage) lies along the Brevard
fault zone, which runs through Randolph and Chambers counties in Alabama
(Medlin and Crawford 1973). The current configuration of Wehadkee Creek as it
crosses the fault suggests stream offset, and perhaps capture, of a nearby Tallapoosa
tributary, via an earthquake or strike slip (Wallace 1968). Many stream channels
show a characteristic sharp jog to the left in these areas, and are high gradient,
which is true of the upper Wehadkee Creek system (Fig. 1; Schulz and Wallace
2013, Wallace 1968). In addition, evidence of the garnet-mica schist found in sections
of the Brevard Fault can be found in abundance at High Shoals, providing
evidence that the underlying topography is part of the fault. The timing of these
geologic events is unknown.
Fish distributions are often difficult to explain in light of human modifications.
However, with 3 species suggesting entries into reciprocal drainages in 1
area, together with geological evidence that suggests conditions were conducive
to stream capture, we believe that such an event was responsible for moving 2
Tallapoosa River endemics into Wehadkee Creek (Chattahoochee River drainage)
and a Chattahoochee River endemic into the Tallapoosa River drainage.
Additional work to estimate timing of the possible transfer, relative to clade
divergence times, would provide a more robust explanation of the possible
historical event or events that have contributed to the observed repeated crossdrainage
distributions of our focal species.
Southeastern Naturalist
125
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
Acknowledgments
We thank Steve Rider, Meagan Roy, and Maria Jarrett for help with field work. We are
grateful to the Belcher family for access to High Shoals, and to Erin Bloom and Mattie
Lewis for assistance with lab analysis.
Literature Cited
Baker, W.H., R.E. Blanton, and C.E. Johnston. 2013. Diversity within the Redeye Bass,
Micropterus coosae (Perciformes: Centrarchidae), species group, with descriptions of
four new species. Zootaxa 4:379–401.
Bishop, P. 1995. Drainage rearrangement by river capture, beheading and diversion. Progress
in Physical Geography 19:449–473.
Boschung, H.T., Jr., and R.L. Mayden. 2004. Fishes of Alabama. Smithsonian Books,
Washington, DC.
Burridge, C.P., D. Craw, and J.M. Waters. 2006. River capture, range expansion, and cladogenesis:
The genetic signature of freshwater vicariance. Evolution 60:10 38–1049.
Connelly, H.M., C.R. Tabit, and L.G. Kral. 2006. Population structure of the Tallapoosa
Shiner (Cyprinella gibbsi) and the Tallapoosa Darter (Etheostoma tallapoosae). Southeastern
Naturalist 5:85–92.
Howard, J.H., and R.P. Morgan. 1993. Allozyme variation in the Mottled Sculpin (Cottus
bairdi): A test of stream-capture hypotheses. Copeia 1993:870–875.
Hurwood, D.A., and J.M. Hughes. 1998. Phylogeography of the freshwater fish Mogurnda
adspersa in streams of northeastern Queensland, Australia: Evidence for altered drainage
patterns. Molecular Ecology 7:1507–1517.
Jenkins, R.E., E.A. Lachner, and F.J. Schwartz. 1972. Fishes of the central Appalachian
drainages: Their distribution and dispersal. Pp. 43–117, In P.C. Holt (Ed.). The distributional
history of the biota of the Southern Appalachians. Part III: Vertebrates.
Resource Division Monograph 4, Virginia Polytechnic Institute and State University,
Blacksburg, VA.
Macenia, M.J., C.E. Johnston, and D.G. Stormer. 2007. Distribution, abundance, biology,
and habitat use of Shoal Bass and species associates in selected tributaries of the Chattahoochee
River, Alabama. Report to Alabama Department of Conservation and Natural
Resources, Montgonery, AL.
Mayden, R.L. 1988. Vicariance biogeography, parsimony, and evolution in North American
freshwater fishes. Systematic Biology 4:329–355.
Medlin, J.H., and T.J. Crawford. 1973. Stratigraphy and structure along the Brevard
fault zone in Western Georgia and Eastern Alabama. American Journal of Science
273A:89–104.
Nylander, J.A.A. 2004. MrModeltest v2. Program. Distributed by the author. Evolutionary
Biology Centre, Uppsala University, Uppsala, Sweden.
Ronquist, F., and J.P. Huelsenbeck. 2003. MrBayes 3: Bayesian phylogenetic inference
under mixed models. Bioinformatics 19:1572–1574.
Schulz, S.S., and R.E. Wallace. 2013. The San Andreas Fault. Available online athttp://
pubs.usgs.gov/gip/earthq3/safaultgip.html. Accessed January 2016.
Strange, R.M. 1998. Mitochondrial DNA variation in Johnny Darters (Pisces: Percidae)
from Eastern Kentucky supports stream capture for the origin of Upper Cumberland
River fishes. American Midland Naturalist 40:96–102.
Wall, B.R. 1968. Studies on fishes of the Bear Creek system of the Tennessee River drainage.
Unpublished M.Sc. Thesis. University of Alabama. Tuscaloosa, AL. 96 pp.
Southeastern Naturalist
A. Jarrett, W. Stiles, A. Janosik, R. Johansen, and C. Johnston
2017 Vol. 16, No. 1
126
Wallace, R.E. 1968. Notes on stream channels offset by the San Andreas Fault, southern
coast ranges, California. Pp. 1–21, In W.R. Dickinson and A. Grantz (Eds.). Proceedings
of Conference on Geologic Problems of San Andreas Fault System. Stanford University
Publications,Geological Sciences, Stanford, CA. Volume 11. 374 pp.
Waters, J.M., D. Craw, J.H. Youngson, and G.P. Wallis. 2001 Genes meet geology: Fish
phylogeographic pattern reflects ancient, rather than modern, drainage connections.
Evolution 55:1844–1851.
Wiley, E.O. 1988. Vicariance biogeography. Annual Review of Ecology and Systematics
19:513–542.