Content uploaded by Martina Medina-Nava
Author content
All content in this area was uploaded by Martina Medina-Nava
Content may be subject to copyright.
Systematics and Biogeography of the Silverside Tribe Menidiini
(Teleostomi: Atherinopsidae) Based on the Mitochondrial ND2 Gene
Devin D. Bloom
1,2
, Kyle R. Piller
1
, John Lyons
3
, Norman Mercado-Silva
4
, and
Martina Medina-Nava
5
The silverside fish tribe Menidiini (Teleostei: Atherinopsidae) consists of four genera, Menidia,Labidesthes,Poblana, and
Chirostoma, that are distributed along the Atlantic coast of North America, throughout the Gulf of Me´xico, insular
United States, and the Mesa Central of Me´xico. It has been suggested that Chirostoma,Poblana, and Menidia should be
recognized as a single genus under the nominal Menidia. To test this hypothesis, phylogenetic relationships within the
tribe Menidiini were assessed using the mitochondrially encoded ND2 gene. Monophyly of the Menidiini tribe was
supported. Results also failed to support monophyly for the genera Menidia and Chirostoma as currently recognized. A
central Mexican clade, inclusive of Chirostoma and Poblana, was recovered as monophyletic and strongly supported.
Relationships within the Mesa Central clade support a previously recognized ‘‘humboldtianum’’ clade and the paraphyly
of Chirostoma with respect to Poblana.
THE New World silverside tribe Menidiini (sensu
Chernoff, 1986a) consists of approximately 32
species and four genera, Labidesthes,Menidia,Po-
blana, and Chirostoma. The monophyly of the tribe has been
supported by morphological (White, 1985; Chernoff, 1986a;
Dyer, 1997) and allozymic studies (Crabtree, 1987), based on
limited taxon sampling. However, a thorough investigation
of relationships within the tribe has not been presented, and
there is uncertainty regarding the taxonomic validity and
the phylogenetic relationships of the genera and species
within Menidiini (Dyer, 1998). This is in part because of
highly variable meristics within and among taxa (Chernoff
et al., 1981; Chernoff, 1982, 1986b; Duggins et al., 1986;
Barriga-Sosa et al., 2002), coupled with an overall lack of
diagnostic morphological characters in some groups (Dyer
and Chernoff, 1996; Dyer, 1998).
Labidesthes is a monotypic genus found in freshwater
throughout the entire Mississippi River and Great Lakes
Basins, as well as along the Gulf coastal plain from Texas to
South Carolina (Lee, 1980; Fig. 1). The genus Menidia
includes seven or eight species that occur along the Atlantic
and Gulf coasts from Maine to Veracruz, Mexico, the
Mississippi drainage as far north as Missouri, and also into
the Florida Keys (Gilbert and Lee, 1980; Fig. 1). Species of
Menidia are generally estuarine and marine fishes, however
some taxa have entirely freshwater populations. Poblana is
endemic to the crater lakes in the eastern central state of
Puebla, Mexico. Four species/subspecies of Poblana have
been described, including P. alchichica,P. letholepis,P.
ferdebueni, and P. alchichica squamata, and each taxon occurs
in a separate lake (Miller et al., 2005). The silverside genus
Chirostoma is the most diverse genus in the tribe and has
been referred to as a species flock endemic to the Mesa
Central (Barbour, 1973a; Barbour and Chernoff, 1984;
Echelle and Echelle, 1984). It is comprised of 18–20
recognized species and includes several subspecies (Barbour,
1973b, 2002). Chirostoma is essentially confined to the Mesa
Central in Mexico, although three taxa, C. jordani,C.
mezquital, and C. humboldtianum, extend beyond this region
(Barbour, 1973b; Miller et al., 2005; Fig. 1).
Previous studies have indicated the necessity for a
comprehensive phylogenetic analysis of silverside tribe
Menidiini. These studies were either conducted prior to
the advent of modern cladistic analyses (Barbour, 1973b;
Johnson, 1974) or they indicate that some groups within
Menidiini may not be monophyletic (Gosline, 1948;
Johnson, 1975; Echelle and Echelle, 1984), or, alternatively,
were broader in scope, investigating higher level relation-
ships and did not include adequate taxon sampling of
Menidiini to address the species and generic level relation-
ships within Menidiini (Chernoff, 1986a; Dyer, 1998).
Additionally, as stated earlier, there is some question
regarding the validity of genera within the tribe Menidiini
(Miller et al., 2005; Nelson, 2006). Therefore, our objectives
were threefold: use mitochondrial DNA (mtDNA) sequence
data to assess the monophyly of the tribe Menidiini, assess
phylogenetic relationships among the genera and species
within Menidiini, and discuss the resulting biogeographical
implications.
MATERIALS AND METHODS
Specimen collection and taxon sampling.—We used specimens
from both subspecies, L. sicculus sicculus and L. sicculus
vanhyningi, for the monotypic Labidesthes. Representatives
of all species of Menidia were included with the exception of
M. c.f. audens and M. clarkhubbsi.Menidia clarkhubbsi is a
gynogenic species from a male M. beryllina and female M.
peninsulae; thus, its mitochondrial genome is identical to M.
peninsulae (Echelle and Mosier, 1981; Echelle et al., 1983,
1989). All four species of Poblana have not been consistently
recognized at the species level, as P. letholepis and P.
squamata have been recognized as subspecies of P. alchichica
1
Southeastern Louisiana University, Department of Biological Sciences, Hammond, Louisiana 70402; E-mail: (KRP) Kyle.Piller@selu.edu. Send
reprint requests to KRP.
2
Present address: University of Toronto, Department of Ecology and Evolutionary Biology, Toronto, ON, Canada, M6H 2P4; E-mail:
devin.bloom@utoronto.ca.
3
University of Wisconsin Zoological Museum, Madison, Wisconsin 53706.
4
Departamento de Ecologı´a Funcional, Instituto de Ecologı´a A.C., Xalapa, Veracruz, Mexico.
5
Universidad Michoacana de San Nicola´ s de Hidalgo, Morelia, Michoaca´n, Me´xico.
Submitted: 10 July 2007. Accepted: 20 January 2009. Associate Editor: D. Buth.
F2009 by the American Society of Ichthyologists and Herpetologists DOI: 10.1643/CI-07-151
Copeia 2009, No. 2, 408–417
(Guerra Magan˜a, 1986; Miller et al., 2005); however, these
taxonomic designations were not based on a published
phylogenetic or taxonomic analysis, thus for our study all
taxa of Poblana were included in the study. All species and
subspecies of Chirostoma (sensu Barbour, 1973b; Miller et al.,
2005) were included in this study with the exception of C.
aculeatum,C. bartoni,C. charari,C. melanoccus, and C.
mezquital because we were unable to obtain specimens of
these species due to their rarity or possible extinctions
(Lyons et al., 1998; Bloom et al., 2008).
Specimens were collected from the wild with standard
seines, cast nets, electrofishing, purchased from commercial
fishermen, or donated by colleagues. Whole specimens or
fin clips were placed in 95%ethanol. The 75 individuals
used in this study included 26 ingroup and five outgroup
species spanning seven genera, three tribes, and both
subfamilies of the silverside family Atherinopsidae (Fig. 2).
When possible, multiple individuals of the same species
were used, and in the case of widespread species, individuals
from multiple populations were included (see Material
Examined). We included 26 species and 71 individuals from
tribe Menidiini, the focus of our study.
DNA extraction, amplification, and sequencing.—Whole geno-
mic DNA was extracted from samples using the DNeasy
tissue kit (Qiagen, Valencia, CA). Sequence data was
generated for the entire mitochondrial encoded NADH
dehydrogenase subunit 2 (ND2) gene (1047 bp) using PCR
amplification primers GLN and ASN from Kocher et al.
(1995). DNA was amplified in 25 ml reactions consisting of
1–4 ml of template DNA, 2.5 ml buffer, 2.5 ml PCR enhancer,
0.25–0.75 ml MgCl
2
,1ml of each primer, 1 ml of dNTPs, and
0.5 ml of Eppendorf (Westbury, NY) Taq polymerase and
sterilized water for the remaining volume. PCR temperature
profile settings were as follows: a touchdown protocol was
used consisting of an initial denaturation at 94uC for 2 min
followed by 5 cycles each of 94uC for 30 sec, 56, 55, 54uC for
Fig. 1. The distribution of the four genera in silverside tribe Menidiini.
Fig. 2. Higher order relationships in New World Silverside family
Atherinopsidae redrawn from Dyer (1997).
Bloom et al.—Systematics of silversides 409
30 sec, 72uC for 1:15 min, 20 cycles of 94uC for 30 sec, 53uC
for 30 sec, 72uC for 1:15 min, and a final extension of 72uC
for 10 min. PCR products were purified using spin columns
(Qiagen) or ExoSAP-IT exonuclease enzyme (USB, Cleve-
land, OH). Sequencing reactions (10 ml) were conducted
using primers MET and TRP from Kocher et al. (1995) and
Bigdye terminator sequencing kit 3.1 (Applied Biosystems,
Foster City, CA) according to temperature profiles recom-
mended by the manufacturer. Samples were then cleaned
with Edge Biospin columns (EdgeBio, Gaithersburg, MD)
and directly sequenced on an ABI 3100 or 3700 DNA
sequencer. All sequences have been deposited on GenBank
(see Material Examined for accession numbers).
Phylogenetic analyses.—The resulting sequences were edited
and aligned manually using Sequencher ver. 4.5 (Gene Codes
Corp., Ann Arbor, MI). Phylogenetic analyses using maxi-
mum parsimony (MP) were conducted using PAUP* (Swof-
ford, 2003). The MP analysis employed heuristic searches
with equal weights for all characters, 1,000 random stepwise
additions with 100 trees saved at each iteration, and tree
bisection and reconnection (TBR) branch swapping. Clade
support was provided by nonparametric bootstrapping using
1,000 pseudoreplicates and 100 heuristic replicates and TBR
branch swapping with ten trees held at each step.
We used ModelTest 3.06 (Posada and Crandall, 1998) to
infer the best model of DNA sequence evolution based on
the Akaike Information Criterion (AIC; Posada and Buckley,
2004), with each of the three codon positions treated as a
separate data partition. Using this model we then imple-
mented a Bayesian analysis (Huelsenbeck et al., 2001) with
ten million Markov chain Monte Carlo generations and
trees saved every 100 generations using MrBayes 3.1.2
(Ronquist and Huelsenbeck, 2003). Four separate runs were
performed, and the log likelihood scores for each run were
plotted against generations to determine the point at which
stationarity was reached. Trees recovered prior to stationar-
ity were discarded as burn-in. Results from the four separate
runs were compared to determine convergence of log-
likelihood values and posterior probabilities. Posterior
probabilities were considered significant when .0.95.
The integrity of various clades or taxonomic groups of
interest was examined quantitatively using two topology
tests. The groups tested were the monophyly of Chirostoma,
Poblana,Menidia (sensu stricto), and the monophyly of
Menidia (sensu Miller et al., 2005). Under maximum
parsimony criteria, the monophyly of each group was
constrained in separate analyses for comparison of tree
length relative to the most parsimonious tree. For the
Bayesian inference (BI), the monophyly of each group was
constrained and the number of post burn-in trees that fit the
constraints was divided by the total number of post burn-in
trees from the BI analysis. The hypothesis was statistically
rejected if 5%or less of the post burn-in trees recovered a
tested relationship.
RESULTS
Sequence alignment was unambiguous, with no insertions
or deletions. Mean nucleotide frequency for all taxa were A
50.216, C50.377, G 50.161, and T 50.246, and there was
no significant difference in nucleotide composition among
taxa (x
2
5188.264, P50.95). There were 538 variable sites,
483 of which were parsimony informative. The maximum
parsimony analysis resulted in 60 equally parsimonious trees
with a score of 2093. A strict consensus tree is shown in
Figure 3 (CI 50.42, RI 50.79, and RC 50.33).
Maximum parsimony and BI methods yielded nearly
congruent topologies with the only exceptions correspond-
ing to the placement of Chirostoma attenuatum and several
minor differences at the tips of the ‘‘Poblana’’ clade. The
monophyly of the silverside tribe Menidiini (sensu Chernoff,
1986a) was strongly supported by both maximum parsimo-
ny and Bayesian inference. At the generic level, Poblana and
Labidesthes were recovered as monophyletic, whereas Meni-
dia and Chirostoma were paraphyletic.
Labidesthes was sister to the remainder of the tribe,
excluding M. extensa. Our data set included two specimens
of L. sicculus vanhyningi from Florida and one individual of L.
s. sicculus from the Upper Mississippi River basin. Average
uncorrected sequence divergence between the two taxa was
14.7%and the specimens of L. s. vanhyningi had identical
haplotypes.
Several species of Menidia formed a monophyletic ‘‘Meni-
dia’’ clade exclusive of M. extensa and M. menidia (Figs. 3, 4).
Menidia beryllina was a monophyletic basal lineage of the
‘‘Menidia’’ clade, while M. peninsulae was paraphyletic with
M. conchorum nested within, and M. colei sister to the M.
peninsulae/conchorum lineage. However, Menidia as recog-
nized traditionally and by Miller et al. (2005) was not
recovered as monophyletic (Figs. 3, 4). Topology tests
indicated that the traditionally recognized Menidia would
require 35 additional steps in MP and was not recovered in
any of the post burn-in trees (0/80,000).
Chirostoma formed a paraphyletic assemblage with C.
riojai,C. arge, and C. contrerasi more closely related to
Poblana than to other members of Chirostoma. The remain-
ing species of Chirostoma formed a large monophyletic
‘‘Chirostoma’’ group of two major clades, the ‘‘humboldtia-
num’’ clade of Barbour (1973b) and a clade with C. jordani
sister to C. larbarcae. All ten individuals from nine
populations of the widespread C. jordani were recovered as
a monophyletic group. Within the ‘‘Chirostoma’’ group MP
placed C. attenuatum as sister to the rest of the ‘‘Chirostoma’’
group, whereas BI placed C. attenuatum as sister to the
‘‘humboldtianum’’ clade within the larger ‘‘Chirostoma’’
group. There was a general lack of resolution within the
‘‘humboldtianum’’ clade, where sequence divergences were
low (approx. 1.0%).
The (C. arge,C. contrerasi)(C. riojai,Poblana) clade forms a
monophyletic group although support values are low, with
an MP bootstrap value of 74%and no support from BI. The
genus Chirostoma was never recovered as monophyletic
among the taxon bipartitions (0/80,000) and required 63
additional steps for monophyly.
The genus Poblana formed a monophyletic group, within a
larger clade that included C. arge,C. contrerasi, and C. riojai to
the exclusion of other species of Chirostoma. The MP analysis
supported monophyly for all of the species of Poblana. A clade
consisting of Poblana ferdebueni was sister to a group
comprising P. alchichica and an unresolved clade inclusive
of P. letholepis and P. squamata. No species of Poblana was
recovered as monophyletic in the Bayesian analysis.
DISCUSSION
Both MP and BI infer a monophyletic Menidiini (sensu
Chernoff, 1986a; Figs. 3, 4), corroborating conclusions
based on morphology (Chernoff, 1986a; Dyer, 1997) and
allozymes (Echelle and Echelle, 1984). Sparks and Smith
410 Copeia 2009, No. 2
Fig. 3. Phylogeny of Menidiini silversides from a strict consensus of 60 equally parsimonious trees resulting following a heuristic search with 1,000
random stepwise additions and 100 trees saved at each iteration. Numbers above branches represent bootstrap values (1,000 pseudoreplicates)
shown when $50.
Bloom et al.—Systematics of silversides 411
Fig. 4. Phylogeny resulting from the Bayesian analysis consisting of 10 million generations. Nodes supported by $95%posterior probabilities are
indicated with an asterisk.
412 Copeia 2009, No. 2
(2004) did not recover Menidiini as monophyletic in a study
of Melanotaenioidei, an Australasian group of atheriniform
fishes. However, they included only four atherinopsid taxa
in a test of monophyly for their ingroup, and thus lacked
taxon sampling of atherinopsids to adequately test relation-
ships of this group.
This study supports the hypothesis that Chirostoma and
Poblana are closely related, comprising a monophyletic
clade that consists of all of the silverside taxa from the
Mesa Central, Mexico. Species of Menidia were basal to a
clade consisting of entirely Mexican taxa (hereafter ‘‘Mex-
ico’’ clade). Species of Menidia are mostly estuarine and
marine in distribution, whereas Chirostoma and Poblana are
exclusively freshwater species suggesting the Mesa Central
taxa likely arose from a single historical transition from
saline to freshwater. Barbour’s (1973b) hypothesis of a
diphyletic origin of Chirostoma also is rejected, although
this study does support the hypothesis that Menidia is the
closest relative to the Mesa Central silversides. Echelle and
Echelle’s (1984) study found that M. beryllina was basal to a
group containing M. peninsulae and all members of Chir-
ostoma, and Poblana (Echelle and Echelle, 1984). Our results
differ in that M. peninsulae and M. beryllina are part of a
monophyletic ‘‘Menidia’’ clade that is sister to a larger clade
including Chirostoma and Poblana.
Labidesthes sicculus was monophyletic and sister to the
remainder of the tribe, with the exception of M. extensa.
Previous studies addressing relationships within Atherinop-
sidae that have included Labidesthes also recovered it as a
basal member of the tribe Menidiini (Echelle and Echelle,
1984; White, 1985; Chernoff, 1986a; Dyer, 1997, 2006).
Florida populations of Labidesthes have been suggested to
represent a distinct species (Bean and Reid, 1930; Grier et al.,
1990; but see Bailey et al., 1954). Although taxonomic
decisions should not be based on sequence divergence
alone, the large degree of sequence variation observed
between populations of L. s. vanhyningi and L. s. sicculus in
this study suggests a more comprehensive examination of
species limits within Labidesthes is warranted.
The phylogenetic placement of Menidia extensa has long
been an enigma (Hubbs and Raney, 1946; Gosline, 1948;
Johnson, 1975; Echelle et al., 1983). Hubbs and Raney (1946)
noted its phylogenetic position was difficult to ascertain as it
shared morphological characters with M. menidia, and also
with M. beryllina. In this study, the basal position of M. extensa
renders Menidia a paraphyletic assemblage. The recovery of a
paraphyletic Menidia brings into question the decision to
treat Chirostoma and Poblana as synonyms of Menidia (Miller
et al., 2005) Miller et al. (2005) based the decision to use an
inclusive Menidia (5Chirostoma +Poblana) on Echelle and
Echelle (1984), a study that did not include M. extensa or M.
menidia. However, the results of our topology test as well as
those of a decay index (not shown) indicate that with only
one additional step M. extensa is no longer the basal taxon in
the tribe. Therefore, we refrain from revising the status of the
genera within the tribe Menidiini until additional individuals
and multiple genes can be added to clarify the taxonomic
status of these groups.
Within the ‘‘Menidia’’ clade, M. beryllina was recovered as
monophyletic, supporting previous hypotheses based on
allozymes that it is a distinct lineage (Johnson, 1975;
Duggins et al., 1986). However, M. peninsulae was not
recovered as monophyletic, with one individual being more
closely related to M. conchorum than to the other specimen
of M. peninsulae. This lack of reciprocal monophyly supports
the conclusion that M. conchorum and M. peninsulae are
conspecific (Duggins et al., 1986). Duggins et al. (1986) and
Johnson (1975) both found M. menidia to be the basal
species of Menidia. The topology within the ‘‘Menidia’’ clade
is nearly identical to that of Echelle et al. (1989) with M.
beryllina as basal to a clade inclusive of M. colei,M.
peninsulae, and M. clarkhubbsi.
The ‘‘Poblana’’ clade from this study is congruent with
Echelle and Echelle (1984), in that C. arge and C. riojai,as
well as Barbour’s (2002) recently described C. contrerasi, are
closely related to Poblana. Both our study and Echelle and
Echelle (1984) suggest that C. arge is the basal taxon,
followed by C. riojai, and that the taxa of Poblana (sensu
stricto) form a monophyletic clade. The species/population
interface continues to be of question among taxa in Poblana.
The parsimony analysis recovered P. alchichica and P.
ferdebueni each as monophyletic lineages, supporting both
at the status of species. Meanwhile P. letholepis and P.
squamata were not recovered as monophyletic lineages but
rather together formed a single clade to the exclusion of a
single specimen of P. letholepis, which is generally in
agreement with Guerra Magan˜as’s (1986) conclusion that
P. letholepis is more closely associated to P. squamata than to
other taxa of Poblana. The lack of reciprocal monophyly
further questions their appropriate taxonomic assignment,
although this may be explained by incomplete lineage
sorting (Avise, 2000; but see Frost and Kluge, 1994; Skinner,
2004). In contrast to the MP results, the Bayesian analysis
did not recover the same topology, but instead found none
of the species of Poblana to be monophyletic.
The MP placement of P. ferbebueni is of interest in relation
to the other taxa of Poblana because it has shield-shaped
scales as do all Chirostoma and most species of Menidia,
unlike the remaining species of Poblana, which have round
or oval shaped scales (Clyde Barbour, pers. comm.). The
basal placement of P. ferdebueni in this study indicates that
there was a character state change from shield- to round-
shaped scales after divergence of P. ferdebueni and all other
taxa of Poblana.
The species of Poblana are restricted to high altitude crater
lakes in the state of Puebla. De Cserna and Alvarez (1995)
and Guerra Magan˜a (1986) suggested that a pre-Pleistocene
lake covered at least part of the area and receded from the
west to the east, resulting in isolated populations. When the
topology of the ‘‘Poblana’’ clade is used to investigate area
relationships, a straightforward west to east pattern emerges
(Fig. 5). The occurrence of the basal taxa, C. arge and C.
contrerasi, in the Lerma-Santiago basin suggests a former
connection to the Valley of Me´xico. A number of other fish
groups support such a connection, including C. humboldtia-
num and C. jordani (Barbour, 1973a) as well as various groups
of goodeids (Webb et al., 2004; Gesundheit and Macı´as
Garcia, 2005; Dominguez-Dominguez et al., 2006) and
cyprinids (Scho¨ nhuth and Doadrio, 2003; Miller et al.,
2005). If the hypothesis of a pre-Pleistocene lake is correct,
then the east–west trend from basal to derived taxa suggests
that the dry lake receded from west to east. However, the
timing of the Lerma–Santiago connection needs further
investigation to determine whether in fact the ancestor of
‘‘Poblana’’ had access to the Me´xico basin prior to the pre-
Pleistocene lake.
The ‘‘Chirostoma’’ clade is a monophyletic group compris-
ing nearly all of species of Chirostoma. This study supports
Bloom et al.—Systematics of silversides 413
Echelle and Echelle’s (1984) ‘‘humboldtianum’’ clade, which
was a monophyletic group of eight species, all of which were
part of Barbour’s (1973b) ‘‘jordani’’ group to the exclusion of
C. jordani.Chirostoma grandocule and C. patzcuaro, which
were not included in the Echelle and Echelle study (1984),
were also found to be members of the ‘‘humboldtianum’’
group. Our study discovers a novel placement for C. jordani
as sister to C. labarcae.Chirostoma jordani is the most
widespread of all species of Chirostoma and along with C.
humboldtianum and C. mezquital is the only species of
Chirostoma to extend beyond the Mesa Central. Ten
individuals of C. jordani from nine populations of C. jordani
were included in this study in order to thoroughly test the
monophyly of the species. A monophyletic C. jordani was
strongly supported by both analyses.
Neither our study nor Echelle and Echelle (1984) support
Barbour’s (1973b) ‘‘arge’’ group. The lack of MP and BI
agreement in placement of C. attenuatum could explain our
inability to recover the ‘‘arge’’ group especially because we
were only able to analyze four of the eight species in
Barbour’s (1973b) ‘‘arge’’ group. Unfortunately the taxon
sampling within this group is unlikely to improve because of
the extreme rarity of the missing taxa (Lyons et al., 1998;
Bloom et al., 2008).
The ‘‘humboldtianum’’ clade is composed of nine currently
recognized species (sensu Barbour, 1973b; and Miller et al.,
2005). These species make up the bulk of what has been
referred to as a ‘‘species flock,’’ although Barbour and
Chernoff (1984) argue that only those taxa in which
‘‘intrinsic mechanisms play the major role in the diversifi-
cation of populations’’ should be considered in terms of a
species flock. Barbour and Chernoff (1984) recognized the
large piscivorous species found in Lake Chapala, C. lucius,C.
sphyraena,andC. promelas, locally known as pescados
blancos, as a monophyletic clade forming a species flock.
Our results did not recover a monophyletic clade of
‘‘pescados blancos.’’ However, this might be an artifact of
low levels of divergence and resulting general lack of
resolution of relationships within the ‘‘humboldtianum’’
clade. Mitochondrial DNA phylogenies effectively are
single-gene trees with limitations in demarcating species in
the face of possible hybridization and lineage sorting (Avise,
2000). Thus, our results do not resolve the issue of
monophyly for the pescado blanco group but rather
emphasize the need for greater taxon sampling and
information from nuclear genes.
There are a number of competing hypotheses regarding
the derivation of Chirostoma and Poblana. Barbour (1973a)
argued that ancestral forms gained access to the Mesa
Central via a Tertiary marine transgression. Alternatively,
Miller and Smith (1986) suggested Chirostoma was derived
from an ancestor of Menidia that followed a ‘‘Plateau track’’
whereby the Rio Grande (5Rio Bravo) was connected to the
Mesa Central. The Plateau track hypothesis is supported by a
number of fish genera such as Ictalurus,Moxostoma, and an
extinct Micropterus that are primarily found in eastern
United States, but are also represented by species on the
Mesa Central (Lee et al., 1980; Miller and Smith, 1986; Miller
et al., 2005). Miller and Smith (1986) suggested the great
diversity in species number of Chirostoma indicates an earlier
connection than that of other taxa following a similar track.
Although Echelle and Echelle (1984) favored the Plateau
track hypothesis, they also suggested the connection may
have been more recent (Plio-Pleistocene) based on molecular
clock estimations. The relatively low level of DNA sequence
divergence found in our study also supports this more recent
connection. The close relationship of Menidia to Mesa
Central silversides (Chirostoma and Poblana) does not rule
out an origin dating to a marine transgression, but the low
levels of genetic divergence observed in both this study and
that of Echelle and Echelle (1984) seems to support a more
recent origin such as a connection between the Mesa
Central and the Rio Grande. The oldest fossil record of
Chirostoma is thought to be Plio-Pleistocene in age (Miller et
al., 2005), further supporting a more recent connection.
Future studies including a molecular clock calibrated using
fossil data may prove informative in investigating the origin
of Mesa Central silversides.
MATERIAL EXAMINED
Institutional abbreviations follow Leviton et al. (1985). The
GenBank accession numbers for ND2 are included for each
specimen.
Atherinella crystallina:Me´xico, Jalisco, El Tecuan Lagoon,
SLU 5105, EF602045.
Atherinella milleri: Honduras, Rio Cangrejal, SLU5104,
EF602046.
Atherinella schultzi:Me´xico, Chiapas, Rio Palenque, SLU
5103, EF602044.
Basilichthys sp.: Peru, Rio Santuario (n52), ANSP 180736,
EF602042–EF602043.
Chirostoma arge:Me´xico, Guanajuato, Rio Laja, SLU 5110,
EF602099.
Chirostoma attenuatum:Me´xico, Michoaca´n, Lago Pa´tz-
cuaro, SLU 5036, EF602082; Me´xico, Michoaca´n, Lago
Zirahue´n, SLU 5036, EF602083.
Chirostoma chapalae:Me´xico, Jalisco, Lago Chapala (n5
2), SLU 5016, EF602075–EF602076.
Chirostoma consocium:Me´xico, Jalisco, Lago Chapala, SLU
5015, EF602077; Me´xico, Jalisco, Lago Chapala, SLU 5023,
EF602077; Me´xico, Michoaca´n, Lago San Juanico (n53),
SLU 5035, EF602079–EF602081.
Chirostoma contrerasi:Me´xico, Michoaca´ n, Lago Negritas,
SLU 5080, EF602098.
Chirostoma estor:Me´xico, Michoaca´ n, Lake Pa´ tzcuaro (n5
2), SLU 5114, EF602067–EF602068; Mexico, Michoaca´n,
Lake Zirahue´n, SLU 5026, EF602067.
Fig. 5. Cladogram of area relationships within the ‘‘Poblana’’ clade
imposed on a map of the Mesa Central, Me` xico.
414 Copeia 2009, No. 2
Chirostoma grandocule:Me´xico, Michoaca´ n, Lake Pa´ tz-
cuaro (n52), SLU 5118, EF602061–EF602062.
Chirostoma humboldtianum:Me´xico, Michoaca´n, Lago
Texpuxtepec, SLU 5039, EF602073; Me´xico, Michoaca´n,
Lake Zacapu (n52), SLU 5011, EF602071–EF602072;
Me´xico, Michoacan, Presa Santa Teresa, SLU 5119,
EF602074; Me´xico, Nayarit, San Pedro Lagunillas, SLU
5095, EF602070.
Chirostoma jordani:Me´xico, Jalisco, Lago Chapala (n52),
SLU 5033, EF602086–EF602087; Me´xico, Jalisco, Lago Ato-
tonilco, SLU 5046, EF602088; Me´xico, Michoaca´n, Lago
Negritas, SLU 5081, EF602089; Me´xico, Michoaca´n, Lake
Cuitzeo, SLU 5111, EF602090; Me´xico, Michoaca´n, Presa
Alvareina, SLU 5030, EF602091; Me´xico, Jalisco, Rio Mazcu´a,
SLU 5044, EF602092; Me´xico, Jalisco, Lago San Pablo de
Naszas, SLU 5045, EF602093; Me´xico, Guanajuato, Presa
Ignacio Allende, SLU 5113, EF602094; Me´xico, Guanajuato,
Lago Yuriria, SLU 5112, EF602095.
Chirostoma labarcae:Me´ xico, Jalisco, Lago Chapala (n52),
SLU 5017, EF602084–EF602085.
Chirostoma lucius:Me´xico, Michoaca´n, Lake Negritas, SLU
5022, EF602059.
Chirostoma patzcuaro:Me´xico, Michoaca´n, Lake Pa´tzcuaro
(n52), SLU 5117, EF602063–EF602064.
Chirostoma promelas:TizapanHatchery,novoucher,
EF602060.
Chirostoma riojai:Me´xico, Me´xico, Lago Guadalupe Victo-
ria (n52), SLU 5079, EF602096–EF602097.
Chirostoma sphyraena:Me´xico, Jalisco, Lake Chapala (n5
2), SLU 5025, EF602065–EF602066.
Labidesthes sicculus: United States, Minnesota, Lake Wi-
nona, SLU 5101, EF602056.
Labidesthes s. vanhyningi: United States, Florida, Pine Log
Creek (n52), SLU 5106, EF602057–EF602058.
Membras martinica: United States, Mississipi, Gulf of
Mexico, SLU 5102, EF602047.
Menidia beryllina: United States, Texas, Rio Grande at del
Rio, SLU 5109, EF602048; United States, Louisiana, Bayou
Lacombe, SLU 5108, EF602049.
Menidia colei:Me´ xico, Yucatan Peninsula, no voucher,
EF602051.
Menidia conchorum: United States, Florida Keys, Grassy
Key, no voucher, EF602052.
Menidia extensa: United States, North Carolina, Lake
Waccamaw, no voucher, EF602055.
Menidia menidia: United States, North Carolina, Wrights-
ville Beach, no voucher, EF602050.
Menidia peninsulae: United States, Florida, Panama City,
SLU 5107, EF602054; United States, Florida, Wabasso,
Indian River, no voucher, EF602053.
Poblana alchichica:Me´xico, Puebla, Lago Alchichica (n5
4), SLU 5034, EF602108–EF602111, EF602115.
Poblana ferdebueni:Me´xico, Puebla, Lago Chignahuapan (n
55), SLU 5028, EF602100–EF602104.
Poblana letholepis:Me´xico, Puebla, Lago Preciosa (n53),
SLU 5116, EF602105–EF602107.
Poblana squamata:Me´xico, Puebla, Lago Quechulac (n5
4), SLU 5115, EF6021120–EF602114, EF602116.
ACKNOWLEDGMENTS
Many individuals assisted in the field acquisition of
specimens or graciously donated them. These include H.
Bart, P. Cochran, M. Sabaj, J. Quattro, H. Perry, J. Anderson,
T. Lankford, Pa´ztcuaro crew, H. Buelna, E. Solorio Ornelas,
W. Matamoros, and M. Adams-Mercado. Others including L.
Zambrano and E. Dı´az-Pardo assisted in various capacities
throughout the duration of this project. We are particularly
grateful for C. Barbour’s invaluable help confirming the
identity of many of the specimens used in this project and
for insightful discussions regarding various aspects of the
project. This project was supported by grants to K. Piller
from the Louisiana Board of Regents and Southeastern
Louisiana University. This study was approved by South-
eastern Louisiana University’s IACUC (Protocol #002).
Permits for the collection of specimens were issued by
SEMARNAT (#11374). This manuscript was submitted as
partial fulfillment of a master’s degree at Southeastern
Louisiana University by D. Bloom.
LITERATURE CITED
Avise, J. C. 2000. Phylogeography: The History and
Formation of Species. Harvard University Press, Cam-
bridge, Massachusetts.
Bailey, R. M., H. E. Winn, and C. L. Smith. 1954. Fishes from
the Escambia River, Alabama and Florida, with ecological
and taxonomic notes. Proceedings of the Academy of
Natural Sciences of Philadelphia 106:109–164.
Barbour, C. D. 1973a. A biogeographical history of
Chirostoma (Pisces: Atherinidae): a species flock from the
Mexican Plateau. Copeia 1973:533–566.
Barbour, C. D. 1973b. The systematics and evolution of
genus Chirostoma Swainson (Pisces, Atherinidae). Tulane
Studies in Zoology and Botany 19:97–141.
Barbour, C. D. 2002. Chirostoma contrerasi (Atherinopsidae,
Menidiinae), a new species from Lago de Chapala, Me´ xico,
p. 22–33. In: Libro Jubilar en Honor al Dr. Salvador
Contreras Balderas. M. d. L. L. Vilano (ed.). Monterrey,
Me´xico.
Barbour, C. D., and B. Chernoff. 1984. Comparative
morphology and morphometrics of the pescado blancos
(genus Chirostoma) from Lake Chapala, Me´xico,
p. 111–127. In: Evolution of Fish Species Flocks. A. A.
Echelle and I. Kornfield (eds.). University of Maine Press,
Orono, Maine.
Barriga-Sosa, I. A., A. L. Ibanez-Aguirre, and J. L. Arre-
dondo-Figueroa. 2002. Morphological and genetic varia-
tion in seven species of the endangered Chirostoma
‘‘humboldtianum species group’’ (Atheriniformes: Atheri-
nopsidae). Revista de Biologı´a Tropical 50:199–216.
Bean, B. A., and E. D. Reid. 1930. On a new species of brook
silverside, Labidesthes vanhyningi, from Florida. Proceed-
ings of the Biological Society of Washington 43:193–194.
Bloom, D. D., K. R. Piller, J. Lyons, and N. Mercado-Silva.
2008. Threatened fishes of the world: Chirostoma aculea-
tum. Environmental Biology of Fishes 82:21–22.
Chernoff, B. 1982. Character variation among populations
and the analysis of biogeography. American Zoologist
22:425–439.
Chernoff, B. 1986a. Phylogenetic relationships and reclas-
sification of menidiine silverside fishes with emphasis on
the tribe Membradini. Proceedings of the Academy of
Natural Sciences of Philadelphia 138:189–249.
Chernoff, B. 1986b. Systematics of American atherinid
fishes of the genus Atherinella. I. The subgenus Atherinella.
Proceedings of the Academy of Natural Sciences of
Philadelphia 138:86–188.
Bloom et al.—Systematics of silversides 415
Chernoff, B., J. V. Conner, and B. F. Charles. 1981.
Systematics of the Menidia beryllina complex (Pisces:
Atherinidae) from the Gulf of Me´xico and its tributaries.
Copeia 1981:319–336.
Crabtree, C. B. 1987. Allozyme evidence for the phyloge-
netic relationships within the silverside subfamily Ather-
inopsinae. Copeia 1987:860–867.
de Cserna, Z., and R. Alvarez. 1995. Quaternary drainage
development in central Me´xico and the threat of an
environmental disaster: a geological appraisal. Environ-
mental and Engineering Geoscience 1:29–34.
Domı´nguez-Domı´nguez, O., I. Doadrio, and G. Pe´rez-
Ponce de Leo´n. 2006. Historical biogeography of some
river basins in central Me´xico evidenced by their goodeine
freshwater fishes: a preliminary hypothesis using second-
ary Brooks parsimony analysis. Journal of Biogeography
33:1437–1447.
Duggins, C. F., A. A. Karlin, K. Relyea, and R. W. Yerger.
1986. Systematics of the Key Silverside, Menidia conchorum,
with comments on other Menidia species (Pisces: Ather-
inidae). Tulane Studies in Zoology and Botany 25:133–150.
Dyer, B. S. 1997. Phylogenetic revision of Atherinopsinae
(Teleostei, Atherinopsidae), with comments on the sys-
tematics of the South American freshwater fish genus
Basilichthys Girard. Miscellaneous Publications, University
of Michigan Museum of Zoology 185:1–64.
Dyer, B. S. 1998. Phylogenetic systematics and historical
biogeography of the Neotropical silverside family Ather-
inopsidae (Teleostei: Atheriniformes), p. 519–536. In:
Phylogeny and Classification of Neotropical Fishes. L. R.
Malabarba, R. E. Reis, R. P. Vari, Z. M. S. Lucena, and
C. A. S. Lucena (eds.). Edipucrs, Porto Alegre, Brazil.
Dyer, B. S. 2006. Systematic revision of the South American
silversides (Teleostei, Atheriniformes). Biocell 30:69–88.
Dyer, B. S., and B. Chernoff. 1996. Phylogenetic relation-
ships among atheriniform fishes (Teleostei: Atherinomor-
pha). Zoological Journal of the Linnean Society 117:1–69.
Echelle, A. A., T. E. Dowling, C. C. Moritz, and W. M.
Brown. 1989. Mitochondrial-DNA diversity and the origin
of the Menidia clarkhubbsi complex of unisexual fishes
(Atherinidae). Evolution 43:984–993.
Echelle, A. A., and A. F. Echelle. 1984. Evolutionary
genetics of a ‘‘species flock:’’ atherinid fishes on the mesa
central of Me´xico, p. 94–109. In: Evolution of Fish Species
Flocks. A. A. Echelle and I. Kornfield (eds.). University of
Maine Press, Orono, Maine.
Echelle, A. A., A. F. Echelle, and C. D. Crozier. 1983.
Evolution of an all-female fish, Menidia clarkhubbsi
(Atherinidae). Evolution 37:772–784.
Echelle, A. A., and D. T. Mosier. 1981. All-female fish: a
cryptic species of Menidia. Science 212:1411–1413.
Frost, D. R., and A. Kluge. 1994. A consideration of
epistemology in systematic biology, with special reference
to species. Cladistics 10:259–294.
Gesundheit, P., and C. Macı´as Garcia. 2005. Biogeografı´a
cladı´stica de la familia Goodeidae (Cyprinodontiformes),
p. 319–338. In: Regionalizacion Biogeogra´fica en Iberoa-
merica y To´ picos Afines. J. Llorente-Bousquets and J. J.
Morrone (eds.). Facultad de Ciencias, UNAM, Me´xico City.
Gilbert, C. R., and D. S. Lee. 1980. Menidia beryllina, p. 557.
In: Atlas of North American Freshwater Fishes. D. S. Lee,
C. R. Gilbert, C. H. Hocutt, R. E. Jenkins, D. E. McAllister,
and J. R. Stauffer (eds.). North Carolina State Museum of
Natural History, Raleigh.
Gosline, W. A. 1948. Speciation in the fishes of the genus
Menidia. Evolution 2:306–313.
Grier, H. J., D. P. Moody, and B. C. Cowell. 1990. Internal
fertilization and sperm morphology in the brook silver-
side, Labidesthes sicculus. Copeia 1990:221–226.
Guerra Magan˜a, C. 1986. Ana´ lisis taxono´mico poblacional
de peces aterı´nidos (Chirostoma yPoblana), de las cuencas
endorreicas del extremo sur del altiplano mexicano. Anales
de la Escuela Nacional de Ciencias Biolo´ gicas 30:81–113.
Hubbs, C. H., and E. C. Raney. 1946. Endemic fish faunas of
Lake Waccamaw, North Carolina. Miscellaneous Publica-
tions, University of Michigan Museum of Zoology 65:5–30.
Huelsenbeck, J. P., F. Ronquist, R. Nielsen, and J. P.
Bollback. 2001. Bayesian inference of phylogeny and its
impact on evolutionary biology. Science 294:2310–2314.
Johnson, M. S. 1974. Comparative geographic variation in
Menidia. Evolution 28:607–619.
Johnson, M. S. 1975. Biochemical systematics of the
Atherinid genus Menidia. Copeia 1975:662–691.
Kocher, T. D., J. A. Conroy, K. R. McKaye, J. R. Stauffer,
and S. F. Lockwood. 1995. Evolution of NADH dehydro-
genase subunit 2 in east African cichlid fish. Molecular
Phylogenetics and Evolution 4:420–432.
Lee, D. S. 1980. Labidesthes sicculus,p.557.In:AtlasofNorth
American Freshwater Fishes. D. S. Lee, C. R. Gilbert, C. H.
Hocutt, R. E. Jenkins, D. E. McAllister, and J. R. Stauffer (eds.).
North Carolina State Museum of Natural History, Raleigh.
Lee, D. S., C. R. Gilbert, C. H. Hocutt, R. E. Jenkins, D. E.
McAllister, and J. R. Stauffer (eds.). 1980. Atlas of North
American Freshwater Fishes. North Carolina State Museum
of Natural History, Raleigh.
Leviton, A. E., R. H. Gibbs, Jr., E. Heal, and C. E. Dawson.
1985. Standards in herpetology and ichthyology: part 1.
Standard symbolic codes for institutional resource collec-
tions in herpetology and ichthyology. Copeia 1985:802–
832.
Lyons, J., G. Gonza´lez-Herna´ ndez, E. Soto-Galera, and M.
Guzma´n-Arroyo. 1998. Decline of freshwater fishes and
fisheries in selected drainages of west-central Mexico.
Fisheries 23:10–18.
Miller, R. R., and M. L. Smith. 1986. Origin and geograpy of
fishes of central Me´xico, p. 487–517. In: The Zoogeogra-
phy of North American Freshwater Fishes. C. H. Hocutt
and E. O. Wiley (eds.). Wiley-Interscience, New York.
Miller, R. R., W. L. Minckley, and S. M. Norris. 2005.
Freshwater Fishes of Me´xico. The University of Chicago
Press, Chicago.
Nelson, J. S. 2006. Fishes of the World. John Wiley and
Sons, Inc., New York.
Posada, D., and T. R. Buckley. 2004. Model selection and
model averaging in phylogenetics: advantages of the AIC
and Bayesian approaches over likelihood ratio tests.
Systematic Biology 53:793–808.
Posada, D., and K. A. Crandall. 1998. Modeltest: testing the
model of DNA substitution. Bioinformatics 14:817–818.
Ronquist, F., and J. P. Huelsenbeck. 2003. MRBAYES 3:
Bayesian phylogenetic inference under mixed models.
Bioinformatics 19:1572–1574.
Scho¨nhuth, S., and I. Doadrio. 2003. Phylogenetic rela-
tionships of Mexican minnows of the genus Notropis
(Actinopterygii, Cyprinidae). Biological Journal of the
Linnaean Society 80:323–337.
Skinner, A. 2004. Hierarchy and monophyly. Cladistics
20:498–500.
416 Copeia 2009, No. 2
Sparks, J. S., and W. L. Smith. 2004. Phylogeny and
biogeography of the Malagasy and Australasian rainbow-
fishes (Teleostei: Melanotaenioidei): Gondwanan vicari-
ance and evolution in freshwater. Molecular Phyloge-
netics and Evolution 33:719–734.
Swofford, D. L. 2003. PAUP*: phylogenetic analysis using
parsimony [*and other methods]. v. 4.0b10. Sinauer
Associates, Sunderland, Massachusetts.
Webb, S. A., J. A. Graves, C. Macı´as-Garcı´a, A. E. Magurran,
D. O. Foighil, and M. G. Ritchie. 2004. Molecular
phylogeny of the livebearing Goodeidae (Cyprinidonti-
formes). Molecular Phylogenetics and Evolution 30:527–
544.
White, B. N. 1985. Evolutionary relationships of the Ather-
inopsinae (Pisces: Atherinidae). Contributions in Science,
Natural History Museum of Los Angeles County 368:1–20.
Bloom et al.—Systematics of silversides 417
