Deep Divergence and Phenotypic Stasis of a Paleogene Relic Species Complex of Bullhead Catfish Within Eastern China

Abstract

Aim
The bullhead catfish Tachysurus argentivittatus , an early‐diverging clade within the genus Tachysurus , exhibits characteristics indicative of an ancient origin and potential cryptic divergence. This study aims to elucidate the evolutionary processes underlying T. argentivittatus s.l. by examining the impacts of complex paleotectonic and drainage rearrangements in East Asia. Additionally, we aim to uncover potential cryptic diversity within this widespread species and explore the mechanisms behind its conserved morphology.

Location
East China.

Taxon
Bullhead catfish Tachysurus argentivittatus (Siluriformes: Bagridae).

Methods
We sampled 302 individuals of the currently recognised T. argentivittatus from nine localities across East China, covering over 5000 km of its latitudinal range. A concatenated dataset of two mitochondrial and four nuclear genes was used for phylogenetic reconstruction using Bayesian inference (BI) and Maximum Likelihood (ML) methods. Divergence times between main lineages of T. argentivittatus s.l. were estimated with a mitochondrial two‐gene concatenated dataset in BEAST v.2.5.2, applying four secondary calibration points. A 3D principal component analysis (PCA) of 26 morphological traits was employed to diagnose molecular operational taxonomic units (OTUs) identified in the phylogenetic analysis.

Results
Molecular phylogenetic analyses revealed deep divergences within T. argentivittatus s.l., dating back to at least the early Miocene, resulting in three geographically isolated lineages. Coupled morphometric and meristic analyses indicate the subtle but consistent phenotypic differences among lineages. Our findings support the hypothesis that T. argentivittatus s.l. represents a species complex comprising three phenotypically similar yet distinct species. While T. argentivittatus s. str. is restricted to the Pearl River basin, populations from the Yangtze and Amur Rivers are recognised as a distinct species, T. mica , previously considered a synonym of T. argentivittatus . The population from Hainan Island represents a novel taxon, which should be formally described in future studies. We also discuss the biogeographical implications of our findings.

Main Conclusions
Time‐calibrated phylogenetic analyses suggest that the ancestor of T. argentivittatus s.l. originated during the Paleogene. The uplift of the Tibetan Plateau since the Oligocene/Miocene boundary, leading to topographic inversion and fluctuations in sea levels, has significantly influenced the dispersal and diversification of the T. argentivittatus complex, including rearrangements and intermittent connections of large East Asian rivers. The reduction in phenotypic variation among these lineages of T. argentivittatus may be explained by phylogenetic niche conservatism and niche tracking.

Full text

1 of 15
Journal of Biogeography, 2025; 0:e15144
https://doi.org/10.1111/jbi.15144
Journal of Biogeography
RESEARCH ARTICLE
Deep Divergence and Phenotypic Stasis of a Paleogene Relic
Species Complex of Bullhead Catfish Within Eastern China
WeihanShao1,2 | XingweiCai2 | JianyongWu3 | E.Zhang1
1Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China | 2Hainan Academy of Ocean and Fisheries Sciences, Haikou, China | 3Nanjing
Institute of Environmental Sciences, Ministry of Ecology and Environment, Nanjing,China
Correspondence: Jianyong Wu (wjy@nies.org) | E. Zhang (zhange@ihb.ac.cn)
Received: 24 February 2025 | Revised: 19 March 2025 | Accepted: 7 April 2025
Funding: This work was supported by the National Natural Science Foundation of China.
Keywords: fluvial evolution| integrative taxonomy| morphological conservatism| phylogenetic niche conservatism| small- bodied catfish
ABSTRACT
Aim: The bullhead catfish Tachysurus argentivittatus, an early-diverging clade within the genus Tachysurus, exhibits charac-
teristics indicative of an ancient origin and potential cryptic divergence. This study aims to elucidate the evolutionary processes
underlying T. argentivittatus s.l. by examining the impacts of complex paleotectonic and drainage rearrangements in East Asia.
Additionally, we aim to uncover potential cryptic diversity within this widespread species and explore the mechanisms behind
its conserved morphology.
Location: East China.
Taxon: Bullhead catfish Tachysurus argentivittatus (Siluriformes: Bagridae).
Methods: We sampled 302 individuals of the currently recognised T. argentivittatus from nine localities across East China,
covering over 5000 km of its latitudinal range. A concatenated dataset of two mitochondrial and four nuclear genes was used for
phylogenetic reconstruction using Bayesian inference (BI) and Maximum Likelihood (ML) methods. Divergence times between
main lineages of T. argentivittatus s.l. were estimated with a mitochondrial two- gene concatenated dataset in BEAST v.2.5.2,
applying four secondary calibration points. A 3D principal component analysis (PCA) of 26 morphological traits was employed
to diagnose molecular operational taxonomic units (OTUs) identified in the phylogenetic analysis.
Results: Molecular phylogenetic analyses revealed deep divergences within T. argentivittatus s.l., dating back to at least the early
Miocene, resulting in three geographically isolated lineages. Coupled morphometric and meristic analyses indicate the subtle
but consistent phenotypic differences among lineages. Our findings support the hypothesis that T. argentivittatus s.l. represents
a species complex comprising three phenotypically similar yet distinct species. While T. argentivittatus s. str. is restricted to the
Pearl River basin, populations from the Yangtze and Amur Rivers are recognised as a distinct species, T. mi c a, previously con-
sidered a synonym of T. argentivittatus. The population from Hainan Island represents a novel taxon, which should be formally
described in future studies. We also discuss the biogeographical implications of our findings.
Main Conclusions: Time- calibrated phylogenetic analyses suggest that the ancestor of T. argentivittatus s.l. originated during
the Paleogene. The uplift of the Tibetan Plateau since the Oligocene/Miocene boundary, leading to topographic inversion and
fluctuations in sea levels, has significantly inf luenced the dispersal and diversification of the T. argentivittatus complex, includ-
ing rearrangements and intermittent connections of large East Asian rivers. The reduction in phenotypic variation among these
lineages of T. argentivittatus may be explained by phylogenetic niche conservatism and niche tracking.
© 2025 John Wiley & Sons Ltd.

2 of 15 Journal of Biogeography, 2025
1 | Introduction
China harbours roughly 1800 freshwater species accounting for
around one- tenth of the total global species diversity, far greater
than that of any neighbouring country (Cao etal.2024). One of
the primary factors contributing to the high species diversity in
the country is its expansive territory, characterised by diverse
landforms and complex geographical features in a three- level
ladder- like topography (Gong et al. 2024). The lowest terrain
ladder, covering the vast plains and hills, is generally below
500 m of elevation (Xu etal.2021). This area lies within eastern
China, spanning from the Amur River in the north to the Red
River in the south. The fish fauna of the region is renowned for
its diversity and endemism (Leroy etal. 2019; He et al. 2020).
The freshwater fish diversity in eastern China is still poorly un-
derstood, as evidenced by recent molecular analyses revealing
cryptic diversity in many widespread species (Yang etal. 2013;
Chen etal.2017; Shao etal.2021). Phenotypic stasis poses a sig-
nificant challenge to the identification of cryptic diversity, given
the over- reliance on traditional morphology- based methods for
specie s recognit ion. Consequent ly, the evolutionar y implications
of tectonic events and drainage evolution may remain obscured
due to the failure to detect cryptic diversity in morphologically
conserved widespread species.
Morphological conservatism, wherein species split with-
out substantial changes in morphology, represents a signif-
icant area of evolutionary research (Fišer et al. 2018; Cerca
et al. 2020). So far, the underlying reasons for morphologi-
cal stasis remain poorly understood (Fišer etal. 2018; Struck
etal. 2018). Recent divergence is commonly believed to con-
tribute to phenotypic crypsis because recently evolved lineages
may not have accumulated sufficient morphological differ-
ences (Renaud et al. 2012; Chenuil et al. 2019; Brownstein
etal.2022). This explanation, however, does not account for all
cases of cryptic speciation and it has been studied that ancient
species can also retain the same or similar phenotypes over
time (Fišer etal.2018; Esquerré etal.2019; Cerca etal.2020;
Natusch etal.2020). Catfishes exhibit low rates of morpholog-
ical evolution possibly owing to reduced selective pressures,
giving rise to pronounced phenotypic conservatism observed
across deeply divergent clades (Egge and Simons2009; Blanton
etal.2013; Stange etal.2018). This makes them ideal model
organisms for exploring the mechanisms underlying morpho-
logical conservatism over extended timescales.
Tachysurus argentivittatus (Regan1905) serves as an exceptional
model for investigating deep divergence and morphological sta-
sis in ancient species. This bagrid catfish represents a Tertiary
relict species possibly tied to the refugial landscapes of eastern
China (He etal.2020). It was an early- diverging lineage within
Tachysurus Lacepède 1803 (Ku etal.2007), with the crown age
of the genus estimated at approximately 43 million years ago
(Ma) (Zeng2013; Kappas etal.2016). Despite its diminutive size,
less than two inches in standard length (SL), making it one of
the smallest members of the Bagridae, T. argentivittatus boasts
an extensive distribution from the Amur River in the north to
Hainan Island in the south, encompassing the entire latitudinal
gradient of eastern China (Chu etal.1999; Ng2009). Within the
species- rich, East Asia- endemic genus Tachysurus, few species
possess such an extensive geographic range as T. argentivittatus.
Generally, small- sized species exhibit low vagility, restricted
geographic ranges and narrow ecological niches, which in-
crease the likelihood of isolation among drainages (Martin and
Bonett 2015; Roxo et al. 2 017). Cryptic diversity within T. ar-
gentivittatus, as a morphology- based species, is implicated in its
wide distribution, small body size and conservative morphology.
This is particularly noteworthy given that T. argentivittatus has
been synonymised with T. mica (Gromov1970), first described
from the Amur River basin in the Russian Far East, due to the
absence of compelling diagnostic traits (Ng2009). These factors
suggest that the current taxonomy of T. argentivittatus requires
thorough molecular investigation.
Additionally, the ongoing uplift of the Tibetan Plateau since the
Cenozoic has caused significant topographic inversion in East
Asia (Zheng et al.2013; Zhang, Cui, et al.2021; Zhang, Daly,
etal.2021), leading to substantial rearrangements of major riv-
ers, such as the Yangtze River (Zhang etal.2016; Fu etal.2020)
and the Pearl River (Cui etal. 2018; Zhang, Cui, et al. 2021;
Zhang, Daly, et al. 2021). Sea- level fluctuations during this
period have also facilitated intermittent river linkages (Jiang
etal.2019; Zahiri etal.2019). The geomorphic histories of these
fluvial systems could significantly influence the dispersal and
variance of freshwater fishes (Wong et al. 2 017; Cassemiro
etal. 2023). Understanding the patterns of cryptic diversity in
widespread species, like T. argentivittatus, is not only crucial for
elucidating processes underlying cryptic speciation but also of-
fers valuable insights into the linkage between species diversifi-
cation and paleo- drainage evolution. Furthermore, a taxonomy
that is reflective of true evolutionary history can allow resource
managers to more accurately manage and protect jeopardised
species. Undiscovered cryptic species may have implications for
captive breeding and the release of at- risk species. For example,
the Chinese giant salamander (Andrias davidianus) comprises
multiple genetic lineages in nature, but there is evidence of ge-
netic homogenisation due to captive breeding (Yan etal.2018).
In this analysis, specimens of T. argentivittatus s.l. were col-
lected from across the majority of its known range, including
localities where no prior genetic data had been obtained. The
objective is to uncover potential cryptic diversity within this
widespread species via an integrative analysis of morphological
and molecular data and explore potential mechanisms behind
the conserved morphology, considering both intrinsic ecological
characteristics and extrinsic abiotic factors. Furthermore, this
study aims to elucidate the evolutionary processes of T. arg ent iv-
ittatus s.l. connected with complex paleotectonic and drainage
rearrangement events in Eastern Asia.
2 | Materials and Methods
2.1 | Taxon Sampling, Specimen Collection
and Preservation
Nine sampling locations throughout most of the historically
known distribution of T. argentivittatus s.l. were selected,
spanning the Yangtze, the Pearl, Amur (Heilong) Rivers
of mainland China and the Nandu River of Hainan Island
(Figure 1, Table 1). Sampling localities were numbered
from one to nine. A total of 302 individuals of the currently

3 of 15
recognised T. argentivittatus were collected. Among them, 252
individuals were used for molecular analysis, 145 for morpho-
metric analysis and meristic counts. Sequences from samples
of 11 other congeneric species were also used in the phyloge-
netic analysis (TableS1).
2.2 | DNA Extraction, Amplification
and Sequencing Analyses
Genomic DNA was extracted from alcohol- preserved fin clips
using the DNeasy Tissue Kit (Qiagen, Beijing, China). Two
mitochondrial genes [cytochrome b (cytB) and cytochrome c
oxidase subunit I (COI)] were amplified. Four independent nu-
clear loci were sequenced from a subset of individuals cover-
ing the range of mitochondrial phylogenetic trees. These loci
correspond to the protein- coding genes ectodermal- neural
cortex 1 (ENC1), glycosyltransferase (Glyt), myosin heavy
chain 6 (myh6) and sterol regulatory element binding protein
2 (sreb2). The information for the primers and thermal cycling
profiles used in this study is provided in TableS2. Amplified
products were purified and utilised for direct cycle sequenc-
ing by Aolinuoke Sequencing Company. Amplified sequences
were archived in the public domain database GenBank
FIGUR E  | Map of sampling locations of Tachysurus argentivittatus s.l. Colours corre spond to the lineages (blue, li neage A; red, linea ge B; yellow,
lineage C).

4 of 15 Journal of Biogeography, 2025
(Table S1). Multiple alignments were prepared with MEGA
7.0 for all sequences, based on the amino acid sequences with
the program MUSCLE (Edgar 2004) with the default setting.
For these nuclear genes, the polymorphic positions for indi-
vidual sequences from nuclear loci were carefully inspected
to ensure correct and consistent identification of double peaks
in heterozygotes. Nuclear gene sequences containing more
than one ambiguous site were resolved with PHASE 2.1.1
(Smith etal. 2005) for which input files were prepared using
SEQPHASE (Flot2010). Three runs were performed for each
locus with default settings. Recombination tests for detecting
the longest non- recombining region for each locus were con-
ducted using IMGC (Woerner etal.2007).
2.3 | Phylogenetic Analyses
COI and CytB gene sequences of each analysed specimen
were concatenated to form a dataset for phylogenetic tree
reconstructions with Bayesian inference (BI) and Maximum
likelihood (ML). The general time- reversible model with in-
variant sites and a gamma distribution variation across sites
(GTR + I + G) was selected as the best- fitting model for the
COI and CytB gene in Jmodeltest2 (Darriba etal.2012) based
on Akaike information criterion (Akaike 1974). Tachysurus
virgatus and T. kyphus were selected as the outgroups. For
BI analyses, three independent runs and four Markov chains
(three heated chains and a single cold chain) utilising the best-
fit models were performed in MRBAYES v 3.1.2 (Ronquist and
Huelsenbeck 2003), starting from a random tree. Each run
was conducted for 20 million generations and sampled every
1000 generations, with the first 25% discarded as burn- in. We
checked for stationarity by ensuring that the average stan-
dard deviation of split frequencies between independent runs
approached 0 and that the potential scale reduction factor
equalled 1. We examined the MCMC samples in TRACER 1.5
to ensure that all chains were sampled from the same target
distribution. ML analyses were implemented in RAXML- VI-
HPC (Stamatakis2006) applying a GTR + I + G model. Nodal
support values were estimated from 1000 nonparametric boot-
strap replicates. Nodes with 95% Bayesian posterior probabil-
ities in the BI analysis and nodes with 75% bootstrap support
in the ML analysis were considered strongly supported. We
also used NETWORK 4.6 (Bandelt etal.1999) to construct a
median- joining network for CytB.
Additionally, BI and ML analyses were employed for the nDNA
loci utilising the identical parameter settings as well. T. trunca-
tus (Regan 1913), T. pratti (Günther 1892), T. taeniatus (Günther
1873) and T. sinensis Lacepède 1803 were selected as outgroups.
The optimal nucleotide substitution models, selected using
the Akaike Information Criterion in MRMODELTEST, were
GTR + I for ENC1, GTR + I + G for myh6 and Glyt and H KY + I
for sreb2.
2.4 | Divergence Time Estimation
The divergence time between main lineages in Tachysurus
argentivittatus s.l. was estimated using the mitochondrial
two- gene concatenated dataset implemented in BEAST v.2.5.2
(Bouckaert et al.2019). Due to the absence of reliable fossils
records for Tachysurus, the analysis used four secondary
calibration points from Zeng (2013): (a) the divergence be-
tween Hemibagrus macropterus (Bleeker, 1870) and Mystus
Scopoli, 1777 + other species of Hemibagrus Bleeker, 1862
[mean = 43.3 Ma], defining a 95% range of 37.8–49.2 Ma, (b) the
crown age of the Bagridae (mean = 57.6 Ma), defining a 95%
range of 48.2–66.4 Ma, (c) the divergence between Tachysurus
and Pseudomystus + Bagrichthys [mean = 42.1 Ma], defin-
ing a 95% range of 35.9–48.2 Ma, (d) the divergence between
Pseudomystus Jayaram, 1968 and Bagrichthys Bleeker, 1857
[mean = 27.7 million years ago (Ma)], defining a 95% range of
20.8–35.2 Ma. The range of divergence times of Hemibagrus
macropterus and Mystus + other species of Hemibagrus is
consistent with the fossil record of Mystus from the Eocene
(34 Ma) (Chang and Zhou1993).
For the calibrated analysis, an initial preliminary run was
performed with 2.5 × 109 MCMC generations, sampling every
5.0 × 104 generations. Based on the results of the preliminary
analysis, the scale factors and sizes of scale operators were sub-
sequently adjusted. A second analysis was conducted with ad-
justed parameters, comprising four independent replicates of
5.0 × 109 MCMC generations each, with sampling every 1.0 × 105
generations. Stationarity, convergence and effective sample
sizes were assessed using Tracer, with 10% of the samples from
each run discarded as burn- in. The species trees from the four
replicates were then merged, and the MCC species tree was an-
notated following the same procedure used for the ultrametric
mtDNA gene tree.
TABLE  | Sample localities for T. argentivittatus s.l. in this study.
Locality numbers correspond to Figure1.
No.
Locality
abbreviation Locality River basin
1HARB Ha'er Bin City,
Heilong jiang
Province
Amour River
2JMS Jiamusi City,
Heilong jiang
Province
Amour River
3YUEY Yueyang City,
Hunan Province
Middle
Yangtze R iver
4NANX Nanxian County,
Hunan Province
Middle
Yangtze R iver
5LIUY Liuyang City,
Hunan province
Middle
Yangtze R iver
6DC Duchang County,
Jiangxi Province
Lower
Yangtze R iver
7NANC Nanchang City,
Jiangxi Province
Lower
Yangtze R iver
8DA Du'an County,
Guangxi Province
Middle
Pearl River
9HAIK Haikou city,
Hainan province
Nandujiang
River

5 of 15
2.5 | Morphology
Methods for counts and measurements follow Cheng
etal.(2008). All measurements are made point- to- point, never
by projections, with a dial calliper and recorded to 0.1 mm. A
total of 26 measurements are made in this study.
Principal component analysis (PCA) has been proven to
be useful in the diagnosis of molecular OTUs (operational
taxonomic units) recovered in the phylogenetic analysis
(Shao etal. 2021). Prior to PCA, we adopted the method of
Reist (1985) to normalise all morphometric measurements
except standard length to eliminate the effect of allometry of
body parts and sample size on the morphometric data. The
formula of the corrected measurements was given as follows:
Mtrans log M- b (log BL- log BLm) where Mtrans is the size-
transformed measurement for each individual; M is the orig-
inal unadjusted measurement; b is the allometric coefficient
that was calculated as the slope of log M against log BL; BL is
the standard length of each individual; and BLm is the over-
all mean standard length of one population while log is the
base 10 logarithm. All measurements except standard length
were transformed separately utilising the regression slope
and common overall mean body length. The 3D- PCA was run
and visualised with MATLAB (Mathworks Inc., New Mexico,
America). Pairwise comparisons were performed using the
Student t- test. Heterogeneity among multiple samples was an-
alysed using one- way ANOVA. All statistical analyses were
performed using SPSS 21.0 (SPSS Inc., Chicago, IL, USA),
and differences were considered statistically significant at
p < 0.0 01.
Vertebral count has been considered as a key character dis-
tinguishing closely related species of Tachysurus in previ-
ous studies (Cheng etal.2021; Shao et al.2021). The counts
of vertebrae are taken from photographs of Micro- CT (X-
ray- based micro- computed tomography) scanning to ex-
plore internal morphology. Micro- CT scanning was made
in a Siemens SOMATOM Definition X- ray machine (120 kV,
400 mA and 0.4 mm slice thickness and voxel dimensions
0.16 mm × 0.16 mm × 0.16 mm). Micro- CT data are imported
into the Mimics medical imaging software (Materialise N.V.,
Leuven, Belgium), and a 3D digital reconstruction of each
block is created.
3 | Results
3.1 | Mitochondrial and Nuclear DNA Tree
Topology
For the mtDNA dataset, after alignment and trimming the
ends, a total of 2554 base pairs (bp) (COI 1467 bp and CytB
1087 bp) after alignment and trimming the ends were ob-
tained for all 252 sampled specimens. Out of these sites in the
combined mtDNA dataset, 349 were variable, and 291 were
parsimony- informative. For all samples of T. argentivittatus
s.l., haplotype diversity (h) and nucleotide diversity (π) were
relatively high (h = 0.935, π = 0.0 46) ( Table 2). The average
pairwise distances between the three lineages (A, B and C in
Figure2) detected for T. argentivittatus s.l. ranged from 5.7%
to 10.9% (Table3). For the nuclear dataset, a total of 3471 bp,
consisting of four independent loci with lengths ranging from
788 bp (myh6) to 977 bp (sreb2), were obtained from 51 sam-
ples of the major mtDNA genealogy of T. argentivittatus s.l.
The number of individuals, the length of each nDNA locus,
the variable sites and the informative parsimony sites are pre-
sented in TableS1. The average pairwise distance of nuclear
loci between the three lineages ranged from 0.5% to 1.4% as
displayed in Table3.
The topologies of the BI and ML trees generated from the
mtDNA haplotypes are identical. The posterior probability (PP)
and ML bootstrap support all samples of T. argentivittatus s.l.
as a well- supported monophyletic group (PP = 1.0, BP = 100)
with three well- supported lineages showing a strong phylogeo-
graphic structure (Figure 2). Lineage A, comprised of samples
from the mid- lower Yangtze River and Amur River basins, ap-
pears as the sister group of lineages B and C. Samples from the
Pearl River basin and Hainan Island are referred to as lineages B
and C respectively. Additionally, samples from the Amur River
form a monophyletic sub- lineage that is embedded in the para-
phyletic entity formed by the samples spanning the mid- lower
Yangtze R iver.
Eighty- nine haplotypes were detected in the combined mtDNA
dataset, of which 80 are unique to one or two individuals, and
nine haplotypes are shared by 6 to 48 individuals. The median-
joining network (MJN) for the mtDNA data shows a similar
structure to the phylogenetic tree and is presented in Figure3.
In lineage A, haplotypes of the Amur River are separated from
those of the mid- lower Yangtze River.
The nDNA phylogenetic trees using four nDNA genes (ENC1,
Glyt, sreb2 and myh6) show that the three mtDNA lineages
form reciprocally independent clusters, although the support is
weak in some nodes (Figure4). Moreover, except for the myh6
gene, the subtle geographic structure of lineage A found in the
mtDNA gene tree is not supported by nDNA datasets as samples
from the mid- lower Yangtze River and Amur River basins are
mixed and clustered together.
3.2 | Divergence Time Estimation
Divergence dating estimated that the most recent common an-
cestor (MRCA) of T. argentivittatus s.l. emerged in the middle
Oligocene (ca 27.5 Ma) (Figure5). Two major cladogenic events
occurred within T. argentivittatus s.l.: Lineage A diverged from
the remaining lineages ca. 17.8 Ma [95% Highest Posterior
Density (HPD): 15.1–21.9 Ma] during the early Miocene. The
TABLE  | Genetic diversity and demographic statistics for T.
argentivittatus s.l. based on mitochondrial markers.
n N h ± SD π ± SD
Clade A 84 62 0.977 0.0026
Clade B 71 11 0.796 0.0021
Clade C 98 16 0.693 0.0013
Total 253 89 0.935 0.046

6 of 15 Journal of Biogeography, 2025
divergence time between lineages B and C was calculated to be
about 7.6 Ma (95% HPD: 5.7–9.9 Ma) in the late Miocene.
3.3 | Morphology
A three- dimensional scatterplot of the three principal compo-
nents (PC1- 3) from linear measurements separated the three lin-
eages uncovered by molecular analysis into three morphospaces
with slight overlap (Figure6). PC1 (49.7%) loaded strongly on the
pre- anal length and body depth, PC2 (11.7%) on maxillary, inner
and outer mandibular barbel length, and PC3 (7.9%) on the head
length and caudal peduncle length. A one- way ANOVA demon-
strated significant differences across all principal components
(p < 0.001), indicating that morphometric characters effectively
distinguish the samples.
The morphometric measurements highlight how morphologi-
cally conser ved these OTUs are. Lineages A and B cannot be dis-
tinguished by any metric traits, and only one and two traditional
indices are effective in differentiating Lineage C from Lineage
A and B, respectively. Lineage C differs from Lineage A in hav-
ing longer maxillary barbels (Figure7a) and from Lineage B in
having longer outer mandibular barbels (Figure7b) and anal- fin
base (Figure7c). Pairwise comparisons using the Student's t- test
further confirmed significant differences among these indices
(p < 0.001). To illustrate data dispersion, we incorporated a box
plot analysis of these indices, displaying the median, quartiles
and outliers (Figure7).
By contrast, the three molecular lineages exhibited notable
variations in colour patterns. Lineages B and C display a black
streak along the leading dorsal fin rays, which is absent in lin-
eage A. Lineage C features a broader black longitudinal stripe
along the flank compared to Lineages A and B (see FigureS1).
Furthermore, these three molecular lineages can be further
distinguished by countable traits as Lineage B differs from the
other two lineages in having more vertebrae numbers (30–32 vs.
27–29) (Figure7).
4 | Discussion
4.1 | Species Delineation and Taxonomic
Implications
This study apparently demonstrates that T. argentivittatus
(Regan 1905), as currently recognised, represents a cryptic
species complex with deep divergence. The genetic distances
FIGUR E  | Phylogenetic tree based on Bayesian inference showing the relationships among Tachysurus argentivittatus s.l. for the mitochondri-
al (concatenated CytB and COI ) genes. Colour coding on branches indicates bootstrap values and posterior probabilities from maximum likelihood
analysis and Bayesian analysis, respectively.
TABLE  | Genetic distances of the mtDNA and nuDNA between
major clades of T. argentivittatus sensu lato estimated in Kimura's two-
parameter model.
Combined
COI and
CytB gene ENC1 Glyt myh6 sreb2
Clade A/
Clade B
0.103 0.013 0.016 0.008 0.011
Clade A/
Clade C
0.109 0.014 0.012 0.007 0.009
Clade B/
Clade C
0.057 0.006 0.006 0.004 0.006

7 of 15
between these lineages, ranging from 5.7% to 10.9% for the con-
catenated CytB and COI genes, are substantially higher than the
species- level distances observed among closely related conge-
neric species, such as 1.5% and 1.4% for the CytB gene between
T. taeniatus and T. aurantiacus (Temminck & Schlegel 1846)
and T. longispinalis (Nguyen 2006) and T. kyphus (Mai 1978),
respectively (Ku etal. 2007; Shao etal.2024) and also the cut-
off value (2.0%) commonly utilised to denote intraspecific vari-
ation (Ward etal.2009). These values are comparable to those
observed in catfishes with cryptic diversity. For example, five
deeply divergent cryptic lineages have been identified in the
Amazonian driftwood auchenipterid catfish Centromochlus ex-
istimatus Meets 1974 s.l., with genetic distances ranging from 4%
to 18% (Cooke etal.2012).
Both mtDNA and nuDNA datasets reveal that the morphology-
defined species T. argentivittatus comprises three genetically
distinct and geographically disjunct lineages (A, B and C), ex-
hibiting deep divergence (Figures2–5). Given that three of the
four nuclear gene trees exhibited the same topology as the mi-
tochondrial gene tree, the low support in the nuclear gene trees
is likely caused by Incomplete Lineage Sorting (ILS), as genetic
sorting may lag behind species divergence during speciation,
particula rly for nuclear genes, which typically have slower muta-
tion rates compared to mitochondrial genes (Cooper etal.2017).
Additionally, the larger effective population size of nuclear DNA
extends the sorting time (Chen etal.2019). Geographic and eco-
logical evidence further supports the hypothesis that introgres-
sion is unlikely. T. argentivittatus s.l. predominantly inhabits
the mainstem rivers and lakes. Lineages A and B are restricted
to the Yangtze and Pearl River basins, respectively, with their
estuaries being geographically distant, making gene exchange
through sea- level regression during glacial periods unlikely.
The only plausible mechanism for such exchange would be river
capture events driven by orogenic processes; however, small to
moderate- scale orogeny primarily affects stream- dwelling spe-
cies (Beltrán- López etal.2021).
Coupled with geographical disparity, multiple lines of evidence
robustly support the delineation of the three distinct lineages as
three putative species. More importantly, subtle but consistent
phenotypic differences among the three distinct lineages pro-
vide additional evidence (Figures 6 and 7). The true species T.
argentivittatus s.s. should be restricted to Lineage B, given that
the species was originally described by Regan (1905) based on
three specimens collected from Canton (presently Guangzhou
City, Guangdong Province, in the lower Pearl River basin). The
Yangtze River and Amur River populations are grouped within
Lineage A, showing marginal genetic differentiation as revealed
by both mitochondrial and myh6 gene trees (Figures 2 and 5).
FIGUR E  | Median- joining network of concatenated cytb and COI sequences. Different colours represent different localities, scaled according to
their frequency in the entire sample. The locality abbreviations are consistent with the information presented in Table1.

8 of 15 Journal of Biogeography, 2025
They should be considered conspecific, provided that the intra-
specific level of differentiation is lower (< 0.5%), and no discern-
ible morphological variations are observed between both. We
concur with Bog utskaya etal.(2008) that T. mic a (Gromov1970),
initially described from the Amur River basin in the Russian Far
East, should be revalidated as a species distinct from T. a r gen -
tivittatus. Due to its conservative external morphology, T. m ica
has long been a junior synonym of T. argentivittatus (Ng2009).
Consistent differences in body colouration are found in this
study between the two species: T. mica lacks the black blotches
in the dorsal fin (see Figure S1), a character well illustrated in
its original description (Gromov1970), but this character is pres-
ent in T. argentivittatus (Regan1905). Additionally, T. mica has
fewer vertebrae when compared to T. argentivittatus (Figure7).
Lineage C, so far restricted only to the Nandu- Jiang of Hainan
Island, deserves specific status, mainly owing to the inter-
lineage genetic divergence (5.7%) observed between it and its sib-
ling lineage B (T. argentivittatus s.s.). It is temporarily referred
to as T. aff. argentivittatus, which is warranted to be formally
described in the future.
4.2 | Speciation and Alluvial Rearrangement
The species diversification of the T. argentivittatus complex is
concordant with the reconstructed history of the Pearl River
basin. In response to the intense uprising of the Tibetan Plateau
beginning at the Oligocene- Miocene boundary, river systems of
East Asia were re- organised owing to the dramatic topographic
reversion facilitating the speciation of freshwater fishes (Zheng
et al. 2013). The estimated divergence time between T. mi c a
and its sister group (17.8 Ma; Figure 2) aligns with the step-
wise westward expansion of the Pearl River, transitioning from
a coastal to a continental- scale drainage system since the late
Oligocene (roughly 24.8 Ma) (Zhang, Cui, et al. 2021; Zhang,
Daly, etal.2021). Notably, geological evidence indicates that the
FIGUR E  | Phylogenetic trees based on Bayesian inference showing the relationships among the lineages for the four nuclear genes. Values on
branches indicate bootstrap values and posterior probabilities from maximum likelihood analysis and Bayesian analysis, respectively.

9 of 15
modern Pearl River basin began forming when the Hongshui- He
and Zuo- Jiang systems merged into the Xi- Jiang, one main trib-
utary of this basin. Subsequently, the Gui- Jiang and Liu- Jiang
systems, formerly part of the paleo- Yangtze River system, were
integrated into the Xi- Jiang during the early Miocene (Zhang,
Cui, etal.2021; Zhang, Daly, etal.2021). Tachysuru s mica, so far
misidentified as T. argentivittatus s.l. in China (Chu etal.1999),
is mainly found in the mid- lower Yangtze River basin, notably
in Dongting and Poyang Lakes. Despite being collected only
from the Hongshui- He in this study, T. argentivittatus s.s. was
historically documented from the mid- lower Pearl River basin
(Zheng 1989). The headwaters of the main affluents of Lake
Dongting are geographically in close proximity to those of the
Gui- Jiang and Liu- Jiang systems. This topological network im-
plies that the common ancestor of the species complex likely
came into existence in the present- day mid- lower reaches of
both the Yangtze River and Pearl River basins and then ex-
panded southward into Hainan Island and northeastward into
the Amur River basin, facilitating species diversification. This
expansion may have been driven by large- scale river piracy and/
or ancient basin connections, the critical biogeographic pro-
cesses that fostered inter- basin bio- dispersals and speciation
(Goodier etal.2011; Van Steenberge etal.2020).
The allopatry of the sister pair T. argentivittatus and T. aff. ar-
gentivittatus suggests the historical distribution of their com-
mon ancestor in the Pearl River basin and Hainan Island. The
widespread ancestral population is likely achieved through the
paleo- river system “Kontum- Ying- Qiong” formed in the early
Oligocene when the Tibetan Plateau started to uplift (Pang
FIGUR E  | Time- calibrated tree based on the mitochondrial (concatenated CytB and COI ) genes.

10 of 15 Journal of Biogeography, 2025
et al. 2009). This ancient fluvial system, with its origins in
central Vietnam, flowed eastwards through southern Hainan
Island (Qiongdongnan Basin) (Lu etal.2024; Qiao etal.2024).
The initial forwards- extending river mouth of the Pearl River
joined this ancient fluvial system, resulting in the inter- basin
connection between the Pearl River and river basins of Hainan
Island during the early Miocene (Shao etal.2018). Due to inten-
sified Asian summer monsoons since 9.5 Ma (Collins etal.2018;
Mathew etal.2020), the late Miocene rise in the sea level of the
South China Sea likely submerged the paleo course of the Pearl
River and disrupted the connection between Hainan Island and
mainland China, ultimately giving rise to the splitting between
T. argentivittatus and T. aff. argentivittatus roughly 7.8 Ma.
Similar paleo- river disruptions have also been observed in adja-
cent regions, for example, the Lufeng- Hanjiang Depression that
ceased to receive sediments from the Pearl River basin in the
Late Miocene (Zhang etal.2012).
The intraspecific genetic divergence of T. mica is subtle, with
the Yangtze River population as its base and the Amur River
population as the most recent branches (Figures2 and 3), sug-
gesting a recent south- to- north colonisation of this species. This
colonisation possibly took place before the late Pleistocene as the
Yellow Sea transitioned from a fluvial/lacustrine environment
to a marine one at 0.8 Ma (Lu etal.2024). Indeed, the Yellow
Sea ever served as a corridor for the inter- basin dispersal of
freshwater fishes within East Asia before the late Pleistocene
when a vast ancient river basin existed in this region (Kong
etal.2006; Zhang etal.2019). Palaeobotanical and lithological
evidence indicates a warming phase began around 6.0 Ma (Sun
and Wang2005) and persisted at least into the early Pleistocene
(Meng etal.2018) in Northeast China including the Amur River
basin. This is indicated by a resurgence of broadleaved decidu-
ous trees, such as Castanea ungeri, Quercus miovariabilis and
Fagus stuxibergii, whose modern counterparts are primarily
distributed in mountainous regions south of the Yangtze River
(Sun and Wang2005), where T. argentivittatus s.l. is now con-
centrated. These findings suggest a warm, humid climate in the
Amur Basin during this period, likely facilitating the northward
expansion of T. mi c a.
Since the late Pleistocene, this inter- basin dispersal has been dis-
rupted by subsidence and the creation of the Yellow Sea (Kong
etal.2006) and channel transitions of the Yangtze River (Zheng
etal.2013) and the Songari River (Qiu etal.1979), which were
triggered by the uplift of the northeastern Tibetan Plateau over
this period (Li etal.2014). As a result, the gene flow between
the two populations of T. mica was impeded, contributing to
their very recent divergence. This cladogenesis is consistent
with observations in other freshwater fishes distributed in
these river basins, such as Tachysurus ussuriensis (Dybowski
1872) (Xu etal.2011), Sarcocheilichthys sciistius (Abbott 1901)
(Li etal.2022) and Squalidus argentatus (Sauvage & Dabry de
Thiersant 1874) (Yang etal.2013).
4.3 | Deep Divergence and Phenotypic Stasis
Despite the deep divergence here observed within the T. argen-
tivittatus complex, phenotypic evolution appears to be highly
FIGUR E  | Three- dimensional principal component analysis (PCA) of the morphometric characteristics of T. argentivittatus s.l. The pie charts
represent the proportion of vertebrae number possessed by the corresponding lineages.

11 of 15
conserved, in stark contrast to the more pronounced morpho-
logical disparities documented in other species complexes of the
genus. Notably, each species pair within this complex exhibits
only two to three morphological differences (Figures6 and 7).
This contrasts with the T. albormarginatus (Rendahl 1928) and
T. nitidus (Sauvage & Dabry de Thiersant 1874) complexes, out of
which major lineages, split during the Plio- Pleistocene, display a
high level of morphological divergence, from four to eight differ-
ences per species pair (Shao etal.2021; Shao and Zhang2023).
Understanding the processes that generate species diversifica-
tion without concurrent morphological changes is crucial for
evolutionar y research and biodiversity conservation (Natusch
etal.2020; Esquerré etal.2021). Three primary explanations for
this phenomenon are recent divergence, morphological conver-
gence and phylogenetic niche conservatism (Taylor etal.2019;
Vega- Sánchez etal.2019; Esquerré etal.2021). In the case of the
T. argentivittatus complex, recent divergence and morphological
convergence are unlikely elaborations, given the phylogenetic
structure. Phylogenetic niche conservatism might explain the
retention of similar morphology over time by driving stabilising
selection. In terms of Fišer et al. (2018), if phylogenetic niche
conservatism is accountable for morphological stasis, three con-
ditions are expected: (1) similar ecological niches among cryptic
taxa, (2) morphological constraints due to selection; and (3) allo-
patric speciation. Members of the T. argentivittatus complex in-
deed share similar habitat preference for deep waters of lowland
river channels or lakes on the basis of observations during our
field works. Morphological differences observed within the spe-
cies complex appear to result from neutral selections under sta-
ble selective pressures, with more emphasis on colour morphs
rather than ecological traits (Figure 7). Colour- morphic varia-
tions in non- sexually dimorphic species have been found to be
strongly correlated with neutral genetic markers and, to a lesser
extent, with selective pressures, because they do not play a role
in mate recognition (Blanton etal. 2013; Morales et al. 2 017).
Traits such as mouth orientation and size, body depth, caudal-
fin shape and size and caudal peduncle dimensions are much
more influenced by ecological pressures (Motta etal.1995) and
seem less inf luential in differentiation within the T. argen tivit-
tatus complex. Phylogenetic niche conservatism is thus a more
plausible explanation for the morphological stasis in the T. a r-
gentivittatus complex.
The origin of the T. argentivittatus complex is estimated to date
back to 27.5 Ma (Figure4), supporting the hypothesis that it rep-
resents a Paleogene relic species. Its origin predates the early up-
lift of the Tib etan Plateau, which began during t he late Oligocene
(Zheng etal.2013; Yu etal.2015). The reported Paleogene fish
fossil, unearthed in the northwestern portion of the Qaidam
Basin of western China, is dated to the early late Oligocene
(Chen and Liu2007; Wang etal.2007; Chang etal.2010). This
fossil species (Tchunglinius tchangii Wang and Wu2015), with
lower vertebrae and small body size, represents a fish species
that dwelled in tropic–subtropic lowland warm waters before
the Tibetan Plateau began to uplift (Wang and Wu2015). This
inference is consistent with Jordan's rule that posits a correlation
between vertebral count in fish and environmental factors such
as temperature (McDowall 2008). Interestingly, these two dis-
tinctive traits are shared by the members of the T. argen t ivit ta-
tus complex and T. trilineatus (Zheng 1979) from eastern China
but not by their congeners (Shao etal.2021), and both taxa oc-
cupy basal positions in the phylogenetic tree of Tachysurus (Ku
etal. 2007; Shao etal. 2021). Similar to the fossil fish species
Tchunglinius tchangii, the ancestor of the T. argentivittatus
FIGUR E  | Relation between (a) maxillary barbel length and pre-
anal length for lineages A and B, (b) anal- fin base length and SL, (c)out-
er mandibular barbel length and HL for lineages B and C.

12 of 15 Journal of Biogeography, 2025
complex likely inhabited the warm lowland waters prevalent in
southern China before the intense uplift of the Tibetan Plateau
at the Oligocene–Miocene boundary. This inference can be fur-
ther supported by the fact that descendant species of the T. ar-
gentivittatus complex mainly survive in the remaining lowland
waters with tropical- subtropical climate, such as the Pearl River
and Yangtze River basins, which constitute the concentrated
distribution area of this species complex.
However, the phased uplift of the Tibetan Plateau since the
Cenozoic has triggered topographic transformations, climatic
fluctuations and river system reconfiguration across East Asia,
posing substantial challenges to the persistence of freshwater
fishes in stable environments (Clark et al. 2004; Zhang, Cui,
etal.2021; Zhang, Daly, etal.2021; Cao etal.2018). Obviously,
the ability in niche tracking, the capability to find new areas
where the favourable habitats occur, is essential for species
to attain niche conservatism and morphological stasis (Cerca
etal.2020). This is the case with the T. argentivittatus complex.
The high dispersal capacity of this species complex is implicated
in the broad distribution of T. mic a from the Yangtze River to
the Amur River, thus indicative of effective niche tracking. Its
relatively small- sized body also implies that it does not require
substantial areas of optimal habitats to survive, which enhances
the likelihood of finding favourable habitats through migration
without marked morphological variations.
Author Contributions
Weihan Shao interpreted analyses and wrote the manuscript. Xingwei
Cai collected and analysed the data. E. Zhang and Jianyong Wu pro-
vided critical feedback on manuscript. All authors contributed to the
writing and gave final approval for publication.
Acknowledgements
This work was partially funded by a grant from the National Natural
Science Foundation of China (32303007), Hainan Province Science
and Technology Special Fund (ZDYF2024SHFZ069). We are grateful
to Zhixian Sun (Institute of Zoology, Chinese Academy of Sciences) for
creating the hand- drawn illustrations, to Huanshan Wang (Museum of
Aquatic Organisms at the Institute of Hydrobiology, IHB) for provid-
ing comparative specimens and to Chao Huang (Australian Museum)
for polishing this manuscript. Moreover, according to relevant Chinese
fisheries regulations, T. argentivittatus s.l. is not designated as a pro-
tected species at any level; therefore, no fieldwork permit was required.
Conflicts of Interest
The authors declare no competing financial interests.
Data Availability Statement
The data that supports the findings of this study are available in the
Supporting Information of this article.
References
Akaike, H. 1974. “A New Look at the Statistical Model Identification.”
IEEE Transactions on Automatic Control 19, no. 6: 716–723. https:// doi.
org/ 10. 1109/ tac. 1974. 1100705.
Bandelt, H. J., P. Forster, and A. Rohl. 1999. “Median- Joining Networks
for Inferring Intraspecific Phylogenies.” Molecular Biology and
Evolution 16, no. 1: 37–48. https:// doi. org/ 10. 1093/ oxfor djour nals. mol-
bev. a 026036.
Beltrán- López, R. G., O. Domínguez- Domínguez, K. R. Piller, H. Mejía-
Mojica, A. F. Mar- Silva, and I. Doadrio. 2021. “Genetic Differentiation
Among Populations of the Blackfin Goodea Goodea atripinnis
(Cyprinodontiformes: Goodeidae): Implications for Its Evolutionary
Hist or y.” Journal of Fish Biology 98, no. 5: 1253–1266. https:// doi. org/
10. 1111/ jfb. 14654 .
Blanton, R. E., L. M. Page, and S. A. Hilber. 2013. “Timing of Clade
Divergence and Discordant Estimates of Genetic and Morphological
Diversity i n the Slender Madtom, Noturus e xilis (Ictaluridae).” Molecular
Phylogenetic s and Evolution 66, no. 3: 679 –693. https:// doi. org/ 10. 1016/j.
ympev. 2012. 10. 022.
Bogutsk aya, N. G., A. M. Naseka , S. V. Shedko, etal. 200 8. “The Fishes of
the Amur River: Updated Check- List and Zoogeography.” Ichthyological
Exploration of Freshwaters 19, no. 4: 301–366. https:// doi. org/ 10. 1093/
icesj ms/ fsn132.
Bouckaert, R ., T. G. Vaughan, J. Barido- Sottani, etal. 2 019. “BEAST 2.5:
An Advanced Software Platform for Bayesian Evolutionar y Analysis.”
PLoS Computational Biology 15, no. 4: e1006650. https:// doi. org/ 10.
1371/ journ al. pcbi. 1006650.
Brownstein, C. D., D. Kim, O. D. Orr, et al. 2022. “Hidden Species
Diversity in an Iconic Living Fossil Vertebrate.” Biology Letters 18, no.
11: 20220395. https:// doi. org/ 10. 1101/ 2022. 07. 25. 500718.
Cao, L., L. Shao, P. Qiao, Z. Zhao, and D. J. J. van Hinsbergen. 2018.
“Early Miocene Birth of Modern Pearl River Recorded Low- Relief, High-
Elevation Surface Formation of SE Tibetan Plateau.” Earth and Planetary
Science Letters 496: 120–131. https:// doi. org/ 10. 1016/j. epsl. 2018. 05. 039.
Cao, L., W. H. Shao, W. J. Yi, and E. Zhang. 2024. “A Review of
Conservation Status of Freshwater Fish Diversity in China.” Journal of
Fish Biology 104, no. 2: 345–364. https:// doi. org/ 10. 1111/ jfb. 15606 .
Cassemiro, F. A., J. S. Albert, A. Antonelli, et al. 2023. “Landscape
Dynamics and Diversification of the Megadiverse South American
Freshwater Fish Fauna.” Proceedings of the National Academy of Sciences
120, no. 2: e2211974120. https:// doi. org / 10. 1073/ pnas. 22119 7412 0 .
Cerca, J., C. Meyer, D. Stateczny, et al. 2020. “Deceleration of
Morphological Evolution in a Cryptic Species Complex and Its Link to
Paleontological Stasis.” Evolution 74, no. 1: 116–131. https:// doi. org/ 10.
1111/ evo. 13884 .
Chang, M . M., D. S. Miao, and N. Wang. 2010. “Ascent With Mod ification:
Fossil Fishes Witnessed Their Own Group's Adaptation to the Uplift of
the Tibetan Plateau During the Late Cenozoic.” In Darwin's Heritage
Toda y, edited by M. Y. Long, H. Y. Gu, and Z. H. Zhou, 60–75. Higher
Education Press.
Chang, M. M., and J. J. Zhou. 1993. “A Brief Survey of the Chinese
Eocene Ichthyofauna.” Kaupia 2: 157–162 .
Chen, G. J., and J. Liu. 2007. “First Fossil Barbin (Cyprinidae, Teleostei)
From Oligocene of Qaidam Basin in Northern Tibetan Plateau.”
Vertebrata PalAsiatica 45, no. 4: 330–341.
Chen, W., Z. Zhong, W. Dai, Q. Fan, and S. He. 2017. “Phylogeographic
Structure, Cryptic Speciation and Demographic History of the
Sharpbelly (Hemiculter leucisculus), a Freshwater Habitat Generalist
From Southern China.” BMC Evolutionary Biology 17, no. 1: 1–3. ht tps ://
doi. org/ 10. 1186/ s1286 2- 017- 1058- 0.
Chen, Z., H. Li, X. Zhai, Y. Zhu, and X. Chen. 2019. “Phylogeography,
Speciation and Demographic History: Contrasting Evidence From
Mitochondrial and Nuclear Markers of the Odorrana graminea
Sensu Lato (Anura, Ranidae) in China.” Molecular Phylogenetics and
Evolution 144, no. 6: 106701. https:// doi. org/ 10. 1016/j. ympev. 2019.
106701.
Cheng, J. L., H. Ishihara, and E. Zhang. 2008. “Pseudobagrus
Brachyrhabdion, a New Catfish (Teleostei: Bagridae) From the Middle

13 of 15
Yangtze River Drainage, South China.” Ichthyological Research 55, no.
2: 112–123. https:// doi. org/ 10. 10 07/ s1022 8- 007- 0020- 3.
Cheng, J. L., W. H. Shao, J. A. Lopez, et al. 2021. “Tachysurus Lani, a
New Catfish Species (Teleostei: Bagridae) From the Pearl River Basin,
South China.” Ichthyological Exploration of Freshwaters 1156: 1–17.
https:// doi. org/ 10. 23788/ IEF- 1156.
Chenuil, A., A. E. Cahill, N. Délémontey, E. du Salliant du Luc, and
H. Fanton. 2019. “Problems and Questions Posed by Cryptic Species:
A Framework to Guide Future Studies.” In In “From Assessing to
Conserving Biodiversit y: Conceptual and Practical Challenges”, edited
by E. Casetta, J. Silva, and D. Vecchi, 77–106. Springer Press.
Chu, X. L., B. S. Zheng, and D. Y. Dai. 1999. FAUNA SINICA Osteic hthys
Siluriformes. Science Press (in Chinese).
Clark, M. K., L. M. Schoenbohm, L. H. Royden, et al. 2004. “Surface
Uplift, Tectonics, and Erosion of Eastern Tibet From Large- Scale
Draina ge Patterns.” Tectoni cs 23, no. 1: TC10 06. https:// doi. org/ 10. 1029/
2002T C001402.
Collins, D. S., A. Avdis, P. A. Allison, H. D. Johnson, J. Hill, and M.
D. Piggott. 2018. “Controls on Tidal Sedimentation and Preservation:
Insights From Numerical Tidal Modelling in the Late Oligocene-
Miocene South China Sea, Southeast Asia.” Sedimentology 65, no. 7:
2468–2505. https:// doi. org/ 10. 1111/ sed. 12474 .
Cooke, G. M., N. L. Chao, and L. B. Beheregaray. 2012. “Five Cr yptic
Species in the Amazonian Catfish Centromochlus existimatus
Identified Based on Biogeographic Predictions and Genetic Data.”
PLoS One 7, no. 11: e48800. https:// doi. org/ 10. 1371/ journ al. pone.
0048800.
Cooper, C. D., K. R. Piller, and T. Y. Chu. 2017. “Let's Jump in: A
Phylogenetic Study of the Great Basin Springfishes and Poolfishes,
Crenichthys and Empetrichthys (Cyprinodontiformes: Goodeidae).”
PLoS One 12, no. 10: e0185425. https:// doi. org/ 10. 1371/ journ al. pone.
0185425.
Cui, Y. C., P. Shao, J. Qiao, et al. 2018. “Upper Miocene- Pliocene
Provenance Evolution of the Central Canyon in Northwestern South
China Sea.” Marine Geophysical Research 4 0, no. 2: 223–235. https:// doi.
org/ 10. 10 07/ s1100 1- 018 - 9359- 2.
Darriba, D., G. L. Taboada, R. Doallo, and D. Posada. 2012. “Jmodeltest
2: More Models, New Heuristics and Parallel Computing.” Nature
Methods 9, no. 8: 772. https:// doi. org/ 10. 1038/ nmeth. 2109.
Edgar, R. C. 2004. “MUSCLE: Multiple Sequence Alignment with High
Accuracy and High Throughput.” Nucleic Acids Research 32, no. 5:
1792–1797. https:// doi. org/ 10. 1093/ nar/ gkh340.
Egge, J. J. D., and A. M. Simons. 2009. “Molecu les, Morphology, Missing
Data and the Phylogenetic Position of a Recently Extinct Madtom
Catfish (Actinopterygii: Ictaluridae).” Zoological Journal of the Linnean
Society 155, no. 1: 60–75. https:// doi. org/ 10. 1111/j. 1096- 3642. 2008.
00432. x.
Esquerré, D., S. C. Donnellan, C. J. Pavón- Vázquez, J. Fenker, and J. S.
Keogh. 2 021. “Phylogeography, Historical Demog raphy and Systematics
of the World's Smallest Pythons (Pythonidae, Antaresia).” Molecular
Phylogenetics and Evolution 161: 107181. https:// doi. org/ 10. 1016/j.
ympe v. 2021. 107181.
Esquerré, D., D. Ramírez- Álvarez, C. J. Pavón- Vázquez, et al. 2019.
“Speciation Across Mountains: Phylogenomics, Species Delimitation
and Taxonomy of the Liolaemus leopardinus Clade (Squamata,
Liolaemidae).” Molecular Phylogenetics and Evolution 139: 106524.
https:// doi. org/ 10. 1016/j. ympev. 2019. 106524.
Fišer, C., C. T. Robinson, and F. Malard. 2018. “Cryptic Species as a
Window Into the Paradigm Shift of the Species Concept.” Molecular
Ecology 27, no. 3: 613–635. https:// doi. org/ 10. 1111/ mec. 14486 .
Flot, J. F. 2010. “Seqphase: A Web Tool for Interconverting Phase Input/
Output Files and Fasta Sequence Alignments.” Molecular Ecology
Resources 10, no. 1: 162–166. https:// doi. org/ 10. 1111/j. 1755- 0998. 2009.
02732. x.
Fu, X., W. Zhu, J. Geng, etal. 2020. “The Present- Day Yangtze R iver
Was Established in the Late Miocene: Evidence From Detrital Zircon
Ages.” Journal of Asian Earth Sciences 205: 104600. https:// doi. org/ 10.
1016/j. jseaes. 2020. 1046 00.
Gong, B., Y. Wang, J. A. Zhang, et al. 2024. “Effects of Altitude on
Thyroid Disorders According to Chinese Three- Rung, Ladder- Like
Topography: National Cross- Sectional Study.” BMC Public Health 24,
no. 1: 26. https:// doi. org/ 10. 1186/ s1288 9- 023- 17569 - 5.
Goodier, S. A., F. P. Cotterill, C. O'Ryan, etal. 2011. “Cryptic Diversity
of African Tigerfish (Genus Hydrocynus) Reveals Palaeogeographic
Signatures of Linked Neogene Geotectonic Events.” PLoS One 6, no. 12:
e28775. https:// doi. org/ 10. 1371/ journ al. pone. 0028775.
Gromov, I. A. 1970. “A New Species of Bagrid Catfish [Mystus mica
Gromov, sp. n. (Pisces, Bagridae)] in the Amur Basin.” Vopr osy
Ikhtiologii 10: 40 0–405 (in Russian).
He, D. K., X. Y. Sui, H. S un, etal. 202 0. “Diversity, Pattern and E cological
Drivers of Freshwater Fish in China and Adjacent Areas.” Reviews in
Fish Biology and Fisheries 30, no. 2: 387–404. https:// doi. org/ 10. 1007/
s1116 0- 020- 09600 - 4.
Jiang, X. L., E. M. Gardner, H. H. Meng, M. Deng, and G. B. Xu. 2019.
“Land Bridges in the Pleistocene Contributed to Flora Assembly on the
Continental Islands of South China: Insights From the Evolutionary
Histor y of Quercus Cha mpionii.” Molecular Phyloge netics and Evolut ion
132: 31–45. https:// doi. org/ 10. 1016/j. ympev. 2018. 11. 021.
Kappas, I ., S. Vittas, C. N. P antzartzi, E . Drosopoulou, and Z. G. S couras.
2016. “A Time- Calibrated Mitogenome Phylogeny of Catfish (Teleostei:
Siluriformes).” PLoS One 11, no. 12: e0166988. https:// doi. org/ 10. 1371/
journ al. pone. 0166988.
Kong, G. S., S . C. Park, H. C. Han, J. H. C hang, and A. Mackensen. 2 006.
“Late Quaternary Paleoenvironmental Changes in the Southeastern
Yellow Sea, Korea.” Quaternary International 144, no. 1: 38–52. http s://
doi. org/ 10. 1016/j. quaint. 2005. 05. 011.
Ku, X. Y., Z. G. Peng, R . Diogo, X. Ku, Z. Peng, and S. He. 200 7. “MtDNA
Phylogeny Provides Evidence of Generic Polyphyleticism for East Asian
Bagrid Catfishes.” Hydrobiologia 579, no. 1: 147–159. https:// doi. org/ 10.
1007/ s1075 0- 006- 0401- z.
Leroy, B., M. S. Dias, E. Giraud, et al. 2019. “Global Biogeographical
Regions of Freshwater Fish Species.” Journal of B iogeography 46, no. 11:
2407–2419. https:// doi. org/ 10. 1111/ jbi. 13674 .
Li, J., X. Fang, C. Song, B. Pan, Y. Ma, and M. Yan. 2014. “Late Miocene
Quaternary Rapid Stepwise Uplift of the NE Tibetan Plateau and Its
Effects on Climatic and Environmental Changes.” Quater nary Research
81, no. 3: 400 –423.
Li, X. J., Z. X. Sun, W. Q. Tang, and Y. H. Zhao. 2022. “Revalidation
and Redescription of Sarcocheilichthys sciistius (Abbott, 1901)
(Cypriniformes: Cyprinidae) From the Northern China.” Zootaxa 5141,
no. 4: 341–357. https:// doi. org/ 10. 11646/ zoota xa. 5141.4. 3.
Lu, Y., Y. Cui, P. Qiao, etal. 2024. “New Understanding on the Interplay
of Longitudinal and Transverse Sediment Dispersal: Combination of
3D Seismic Profile and U- Pb Ages of Detrital Zircons of the Northern
South China S ea Paleogene Sedimentar y Rocks.” Journal of Asian Earth
Sciences 275: 106290. https:// doi. org/ 10. 1016/j. jseaes. 2024. 106290.
Martin, S. D., and R. M. Bonett. 2015. “Biogeography and Divergent
Patterns of Body Size Disparification in North American Minnows.”
Molecular Phylogenetics and Evolution 93: 17–28. https:// doi. org/ 10.
1016/j. ympev. 2015. 07. 006.
Mathew, M., A. Makhankova, D. Menier, B. Sautter, C. Betzler, and B.
Pierson. 2020. “The Emergence of Miocene Reefs in South China Sea
and Its Resilient Adaptability Under Varying Eustatic, Climatic and
Oceanographic Conditions.” Scientific Reports 10, no. 1: 7141.

14 of 15 Journal of Biogeography, 2025
McDowall, R. M. 20 08. “Jordan's and Other Ecogeographical Rules,
and the Vertebral Number in Fishes.” Journal of Biogeography 35, no. 3:
501–508. https:// doi. org/ 10. 1111/j. 1365- 2699. 20 07. 01823. x.
Meng, X., L. Liu, X. T. Wang, W. Balsam, J. Chen, and J. Ji. 2018.
“Mineralogical Evidence of Reduced East Asian Summer Monsoon
Rainfall on the Chinese Loess Plateau During the Early Pleistocene
Interglacials.” Earth and Planetary Science Letters 486: 61–69. https: //
doi. org/ 10. 1016/j. epsl. 2017. 12. 048.
Morales, H. E., A. Pavlova, P. Sunnucks, et al. 2017. “Neutral and
Selective Drivers of Colour Evolution in a Widespread Australian
Passerine.” Journal of Biogeography 44, no. 3: 522–536. https:// doi. org/
10. 1111/ jbi. 12942 .
Motta, P. J., K. B. Clifton, P. Hernandez, and B. T. Eggold. 1995.
“Ecomorphological Correlates in Ten Species of Subtropical Seagrass
Fishes: Diet and Microhabitat Utilization.” Environmental Biology of
Fishes 44, no. 1- 3: 37–60. https:// doi. org/ 10. 1007/ BF000 05906 .
Natusch, D. J., D. Esquer ré, J. A. Lyons, etal. 2020. “ Species Delimitat ion
and Systematics of the Green Pythons (Morelia viridis Complex) of
Melanesia and Australia.” Molecular Phylogenetics and Evolution 142:
106640. https:// doi. org/ 10. 1016/j. ympev. 2019. 106640.
Ng, H. H. 2009. “Tachysurus Spilotus, a New Species of Catfish From
Central Vietnam (Teleostei: Bagridae).” Zootaxa 2283: 16–28. ht tps: //
doi. org/ 10. 5281/ zenodo. 191228.
Pang, X. , C. Chen, M. Zhu, etal. 200 9. “Baiyu n Movement: A Significant
Tectonic Event on Oligocene/Miocene Boundary in the Northern South
China Sea and Its Regional Implications.” Journal of Earth Science 20,
no. 1: 49–56. https:// doi. org/ 10. 1007/ s1258 3- 009- 0005- 4.
Qiao, P., Y. Cui, Q. Ma, Q. Yu, and L. Shao. 2024. “Paleogene
Transgression Process and Environmental Evolution in the Deepwater
Area of the Baiyun Depression in the Northern South China Sea.” Acta
Oceanologica Sinica 43, no. 4: 15–24. https:// doi. org/ 10. 1007/ s1313
1- 024- 2340- 9.
Qiu, S. W., G. Y. Sun, W. D. Li, etal. 1979. “The Discovery of the Fossil
Hydrographic Net of Songhua- Jiang in the Three- River Plain.” Acta
Geographica Sinica 34, no. 3: 265–274 (in Chinese).
Regan, C. T. 1905. “Description de six poissons nouveaux faisant partie
de la collection du Musée d'Histoire Naturelle de Genève.” Revue Suisse
de Zoologie 13: 389–393.
Reist, J. D. 1985. “An Empirical Evaluation of Several Univariate
Methods That Adjust for Size Variation in Morphometric Data.”
Canadian Journal of Zoology 63, no. 6: 1429–1439. https:// doi. org/ 10.
1139/ z85- 213.
Renaud, A. K., J. Savage, and S. J. Adamowicz. 2012. “DNA Barcoding
of Northern Nearctic Muscidae (Diptera) Reveals High Correspondence
Between Morphological and Molecular Species Limits.” BMC Ecology
12, no. 1: 1–15. https:// doi. org/ 10. 1186/ 1472- 6785- 12- 24.
Ronquist, F., and J. P. Huelsenbeck. 2003. “MrBayes 3: Bayesian
Phylogenetic Inference Under Mi xed Models.” Bioinformatics 19, no. 12:
1572–1574.
Roxo, F. F., N. K. Lujan, V. A. Tagliacollo, etal. 2017. “Shift Fr om Slow- To
Fast- Water Habitats Accelerates Lineage and Phenotype Evolution
in a Clade of Neotropical Suckermouth Catfishes (Loricariidae:
Hypoptopomatinae).” PLoS One 12, no. 6: e0178240. https:// doi. org/ 10.
1371/ journ al. pone. 0178240.
Shao, L., Y. C. Cui, K. Stattegger, et al. 2018. “Drainage Control of
Eocene to Miocene Sedimentary Records in the Southeastern Margin
of Eurasian Plate.” GSA Bulletin 131, no. 3–4: 461–478. https:// doi. org/
10. 1130/ b32053. 1.
Shao, W. H., J. L. Cheng, and E. Zhang. 2021. “Eight in One: Hidden
Diversity of the Bagrid Catfish Tachysurus albomarginatus s.l.
(Rendhal, 1928) Widespread in Lowlands of South China.” Frontiers in
Genetics 12: 2195. https:// doi. org/ 10. 3389/ fgene. 2021. 713793.
Shao, W. H., J. L. Cheng, and E. Zhang. 2024. “Tachysurus taeniatus
(Günther, 1873), a Senior Synonym of the Congeneric Species T. Ondon
(Shaw, 1934) (Teleostei, Bagridae) From Eastern China.” ZooKeys 1218:
131–144. https:// doi. org/ 10. 3897/ zooke ys. 1218. 135630.
Shao, W. H., and E. Zhang. 2023. “Tachysurus latifrontalis, a New
Bagrid Species From the Jiulong- Jiang Basin in Fujian Province, South
China (Teleostei: Bagridae).” Ichthyological Research 70, no. 1: 110–122.
https:// doi. org/ 10. 1007/ s1022 8- 022- 0 0867 - 0.
Smith, S. A., P. R. Stephens, and J. J. Wiens. 2005. “Replicate Patterns of
Species Richness, Historical Biogeography, and Phylogeny in Holarctic
Treefrogs.” Evolution 59, no. 11: 2433–2450.
Stamatakis, A. 2006. “RAx ML- V I- HPC: Maximum Likelihood- Based
Phylogenetic Analyses With Thousands of Taxa and Mixed Models.”
Bioinformatics 22, no. 21: 2688–2690. https:// doi. org/ 10. 1093/ bioin
forma tics/ btl446.
Stange, M., G. Aguirre- Fernández, and W. Salzburger. 2018. “Study
of Morphological Variation of Northern Neotropical Ariidae Reveals
Conservatism Despite Macrohabitat Transitions.” BMC Evolutionary
Biology 18, no. 1: 1–12. https:// doi. org/ 10. 1186/ s1286 2- 018- 1152- y.
Struck, T. H., J. L. Feder, M. Bendiksby, etal. 2018. “Finding Evolutionary
Processes Hidden in Cryptic Species.” Trends in Ecolog y & Evolution 33,
no. 3: 153–163. https:// doi. org/ 10. 1016/j. tree. 2017. 11. 007.
Sun, X., and P. Wang. 2005. “How Old Is the Asian Monsoon
System?—Palaeobotanical Records From China.” Palaeogeography,
Palaeoclimatology, Palaeoecology 222, no. 3–4: 181–222. https:// doi. org/
10. 1016/j. palaeo. 2005. 03. 005.
Taylor, R. S., M. Bolton, A. Beard, et al. 2019. “Cryptic Species and
Independent Origins of Allochronic Populations Within a Seabird
Species Complex (Hydrobates spp.).” Molecular Phylogenetics and
Evolution 139: 106552. https:// doi. org/ 10. 1016/j. ympev. 2019. 106552.
Van Steenberge, M. W., M. P. Vanhove, A. Chocha Manda, etal. 2020.
“Unravelling the Evolution of Africa's Drainage Basins Through a
Widespread Freshwater Fish, the African Sharptooth Catfish Clarias
gariepinus.” Journal of Biogeography 47, no. 8: 1739–1754. https:// doi.
org/ 10. 1111/ jbi. 13858 .
Vega- Sánchez, Y. M., L. F. Mendoza- Cuenca, and A. González-
Rodríguez. 2019. “Complex Evolutionar y History of the American
Rubyspot Damselfly, Hetaerina americana (Odonata): Evidence of
Cry ptic Speciation.” Molec ular Phylogenetics and Evolution 139: 106536.
https:// doi. org/ 10. 1016/j. ympev. 2019. 106536.
Wang, N., and F. Wu. 2015. “New Oligocene Cyprinid in the Central
Tibetan Plateau Documents the Pre- Uplift Tropical Lowlands.”
Ichthyological Research 62, no. 3: 274–285. https:// doi. org / 10. 1007/ s1022
8- 014- 0438- 3.
Wang, X. M., Z. D. Qiu, Q. Li, et al. 2007. “Vertebrate Paleontology,
Biostratig raphy, Geochronology, and Paleo environment of Qaida m Basin
in Northern Tibetan Plateau.” Palaeogeography, Palaeoclimatology,
Palaeoecology 254, no. 3- 4: 363–385. https:// doi. org/ 10. 1016/j. palaeo.
2007. 06. 007.
Ward, R. D., R. Hanner, and P. D. N. Hebert. 2009. “The Campaign to
DNA Barcode all Fishes, FISH- BOL.” Journal of Fish Biology 74, no. 2:
329–356. https:// doi. org/ 10. 1111/j. 1095- 8649. 2008. 02080. x.
Woerner, A. E., M. P. Cox, and M. F. Hammer. 2007. “Recombination-
Filtered Genomic Datasets by Information Maximization.” Bioinformatics
23, no. 4: 1851–1853. https:// doi. org/ 10. 1093/ bioin forma tics/ btm253.
Wong, W. Y., K. Y. M, L. M. Tsang, and K. H. Chu. 2017. “Genetic Legacy
of Tertiary Climatic Change: A Case Study of Two Freshwater Loaches,
Schistura Fasciolata and Pseudogastromyzon myersi, in Hong Kong.”
Heredit y 119, no. 5: 360–370. https:// doi. org/ 10. 1038/ hdy. 2017. 47.
Xu, H., B. Qi, H. W. Paerl, et al. 2021. “Environmental Controls of
Harmful Cyanobacterial Blooms in Chinese Inland Waters.” Harmful
Algae 110: 102127. https:// doi. org/ 10. 1016/j. hal. 2022. 102190.

15 of 15
Xu, L. F., H. Z. Huang, W. S. Fan, H. M. He, and Y. H. Jia. 2 011. “Genetic
Diversity Analyses of Wild and Cultured Pseudobagrus ussuriensis
Populations.” Marine Sciences 35: 17–22.
Yan, F., J. Lü, B. Zhang, et al. 2018. “The Chinese Giant Salamander
Exemplifies the Hidden Extinction of Cryptic Species.” Current Biology
28, no. 10: R590–R592. https:// doi. org/ 10. 1016/j. cub. 2018. 04. 004.
Yang, J. Q., W. Q. Tang, Y. Sun, etal. 2013. “Microsatellite Diversity
and Population Genetic Structure of Squalidus argentatus (Cypr inidae)
on the Island of Hainan and Mainland China.” Biochemical Systematics
and Ecolog y 50: 7–15. https:// doi. org/ 10. 1016/j. bse. 2013. 03. 023.
Yu, Y., C. Zhang, S. Li, etal. 2015. “Influences of Tibetan Plateau Uplift
on Provenance Evolution of the Paleo- Pearl River.” Chinese Journal of
Geochemistry 34: 208–218.
Zahiri, R., B. C. Schmidt, A. Schintlmeister, R. V. Yakovlev, and M.
Rindoš. 2019. “Global Phylogeography Reveals the Origin and the
Evolutionary History of the Gypsy Moth (Lepidoptera, Erebidae).”
Molecular Phylogenetics and Evolution 137: 1–3. https:// doi. org/ 10.
1016/j. ympev. 2019. 04. 021.
Zeng, Q. 2013. “Molecular Phylogeny and Divergence Time Estimation
of the Bagrid Catfishes (Actinopterygii: Siluriformes).” Master's thesis,
Southwest University.
Zhang, H., Y. Cui, P. Qiao, M. Zhao, and X. Xiang. 2021. “Evolution
of the Pearl River and Its Implication for East Asian Continental
Landscape Reversion.” Acta Geologica Sinica 95: 66–76. https:// doi. org/
10. 1016/j. ympev. 2019. 04. 021.
Zhang, J., S. Wan, P. D. Clift, etal. 2019. “History of Yellow River and
Yangtze River Delivering Sediment to the Yellow Sea Since 3.5 Ma:
Tectonic or Climate Forcing?” Quaternary Science Reviews 216: 74–88.
https:// doi. org/ 10. 1016/j. quasc irev. 2019. 06. 002.
Zhang, X. T., L. Chen, Q. H. She, etal. 2012. “Provenance Evolution of
the Paleo- Hanjian River in the North South China Se a.” Marine Geology
& Quaternary Geology 32, no. 4: 41–48. https:// doi. org/ 10. 372 4/ SP.J.
1140. 2012. 04041 .
Zhang, Z., J. S. Daly, Y. Yan, etal. 2021. “No Connection Between the
Yangtze and Red R ivers Since the Late Eocene.” Marine and Petroleum
Geology 129: 105115. https:// doi. org/ 10. 1016/j. marpe tgeo. 2021. 105115.
Zhang, Z., S. Tyrrell, C. A. Li, etal. 2016. “Provenance of Detrital K-
Feldspar in Jiang han Basin Sheds Ne w Light on the Pliocene- Pleistocene
Evolution of the Yangtze River.” Geological Societ y of America Bulletin
128, no. 9–10: 1338–1351. https:// doi. org/ 10. 1130/ B31445. 1.
Zheng, C. Y. 1989. The Fishes of the Pearl River. Science Press (in
Chinese).
Zheng, H., P. D. Clift, P. Wang, et al. 2013. “Pre- Miocene Birth of the
Yangtze River.” Proceedings of th e National Academy of Sciences 110, no.
19: 7556–7561. https:// doi. org/ 10. 1073/ pnas. 12162 41110 .
Supporting Information
Additional supporting information can be found online in the
Supporting Information section.