Population genomics of Gymnogeophagus labiatus and G. lacustris reinforces the role of hydrographic isolation between evolutionary lineages (Cichlidae: Geophagini)

Pedro Ivo C. C. Figueiredo¹, Diogo T. Hashimoto², Raquel B. Ariede², André L. S. Zani¹, Maikel Varal¹, Luiz R. Malabarba^3,4 and Nelson J. R. Fagundes^1,3,5

PDF: EN    XML: EN | Supplementary: S1 S2 S3 S4 S5 S6 | Cite this article

Abstract​

---

Introduction​

---

Assigning individuals into species’ categories is still one of the most debated issues in the biological sciences. One of the main reasons is the lack of a single concept that could define “species” and delimit these taxa objectively, consistently, and in a biologically meaningful way for all living beings (Hey, 2001; De Queiroz, 2007; Zachos, 2018). In the case of freshwater fishes, while the geographical isolation imposed by different drainages constitutes obvious barriers to gene flow that might align with different evolutionary lineages (or species), shared morphological features may indicate that the same species occurs in different isolated drainages. However, one must be aware that morphological convergence and phenotypic plasticity can represent challenges to the proper identification of taxonomic species, as the former makes similar (but convergent) morphologies to be present in different drainages, and the latter results in different morphologies within the same genealogical lineage (see Figueiredo et al., 2021). These issues may be even more complicated in recent timescales, because it is difficult to distinguish between phenotypic plasticity and incipient speciation in ecological gradients, originating the so-called gray zones of speciation (De Queiroz, 2007; Roux et al., 2016)

Cichlidae is one of the most diverse vertebrate families in the world, famous for being extremely diverse in terms of body shape and coloration pattern, presenting extraordinary cases of evolutionary parallelism in ecological morphologies. Convergent phenotypic evolution is generally interpreted as a strong indication of adaptive evolution (Burress, 2015). Parallelism has already been documented in the main adaptive radiations of African and Neotropical cichlids (Henning, Meyer, 2014) and involves features with a known adaptive function, such as hypertrophic lips (which facilitate larval foraging in rocky crevices) and traits under sexual selection (e.g., body coloration; Urban et al., 2022) and lip size itself (Machado-Schiaffino et al., 2014). Interestingly, it has been shown that lip size evolution in cichlids depends on either genetic (Masonick et al., 2022) and environmental components (Machado-Schiaffino et al., 2014).

There are currently 1,747 valid species in this family, 563 of which belong to the Cichlinae subfamily of Neotropical fish (Fricke et al., 2024). The tribe Geophagini, which forms a clade within Cichlinae, represents approximately 18 genera and 250 species distributed across South America and southern Panama (López-Fernández et al., 2010). Among these, Gymnogeophagus Miranda Ribeiro, 1918 contains 20 valid species (Turcati et al., 2018; Alonso et al., 2019) whose synapomorphies are the presence of a forward-directed spine at the top of the first dorsal pterygiophore and the lack of supraneurals, unique among Neotropical cichlids (Reis, Malabarba, 1988). Representatives of this genus are distributed in Uruguay, Paraguay, Argentina, and in the Brazilian states of Santa Catarina and Rio Grande do Sul (Malabarba et al., 2015; Loureiro et al., 2016; Casciotta et al., 2017), occurring in a series of distinct environments, with varying degrees of ecological specialization.

In the Tramandaí-Mampituba Ecoregion (Ecorregião Tramandaí-Mampituba, in Portuguese, henceforth ETM) (Abell et al., 2008) Reis, Malabarba (1988) recognized two species with distinct ecological habits: Gymnogeophagus labiatus (Hensel, 1870), and Gymnogeophagus lacustris Reis & Malabarba, 1988. ETM has an extensive area covered with diverse types of water bodies (freshwater lakes, lagoons, estuaries, and rivers) that can be divided into two main sub-regions: the rivers and streams of the Serra Geral slope, and the rivers and lagoons of the Coastal Plain of Rio Grande do Sul (PCRS). While the former sub-region occupies deep valleys eroded by basaltic flows dating back some 133 million years ago (mya) (Renne et al., 1992), the latter was formed over the last 400,000 years (Villwock et al., 1984; Tomazelli, Villwock, 2000) and includes the mouth of rio Maquiné at lagoa dos Quadros, the mouth of rio Três Forquilhas at lagoa Itapeva, a series of interconnected lagoons, rio Tramandaí, isolated lakes, wetlands and temporary flood fields (Malabarba, Isaia, 1992). Most lakes in ETM are younger than the Last-Glacial Maximum and found its current configuration only in the Holocene (Tomazelli, Villwock, 2000). The high biological diversity of species observed in the ETM corresponds to around a quarter of the freshwater fish species found in the entire state of Rio Grande do Sul. This high species’ richness is possibly related to the diversity of environments available as well as the distinct historical origin of its components (Langeani et al., 2009; Malabarba et al., 2013).

On the other hand, in the adjacent Patos Lagoon Ecoregion (Ecorregião Laguna dos Patos, in Portuguese, henceforth ELP) (Abell et al., 2008), there is only G. labiatus (Reis, Malabarba, 1988). The ELP represents a much larger (compared to ETM) drainage flowing into the Atlantic Ocean, composed by a complex of 10 river basins in Brazil plus the tributaries of laguna Mirim (or Merín) drainage in Uruguay. This complex includes laguna dos Patos itself, the tributaries that flow directly into the lagoon, such as rio Camaquã, lago Guaíba, rio Jacuí and its tributaries, and laguna Mirim, irrigating a considerable portion of Rio Grande do Sul, as well as the northeast of Uruguay (Becker et al., 2013).

Both species have marked sexual dimorphism, and before the reproductive season the males develop a gibbosity, also known as nuptial (or nuchal) hump, in the anterior region of the head, used for display during the reproductive period or as a fat reserve for the period of caring for the eggs and fry (Lowe-McConnell, 1999). Among the characters described in the literature to differentiate the species, the hypertrophied lip of G. labiatus stands out. Another major difference refers to the coloration characteristics. G. lacustris has a few bright blue spots on the cheek, usually aligned with the series of infraorbital bones, the lip of mature males in the nuptial period is markedly dark orange and the anal fin sometimes has rounded spots. G. labiatus also has bright blue spots on the cheeks, but in greater quantity than those observed in G. lacustris, often aligned in the infraorbital series, and most of the scales on the flank with bright blue spots (Reis, Malabarba, 1988). Gymnogeophagus gymnogenys (Hensel, 1870) is the only other mouth brooder species of the genus that occurs sympatrically with both G. labiatus and G. lacustris. It is not, however, closely related to the clade formed by those two species. Instead, it is deeply inserted in a large clade containing all the 15 remaining mouth brooder species of the genus (Říčan et al., 2019).

Recently, morphological and genetic data have questioned how accurately the current taxonomy reflects the evolutionary history of G. labiatus and G. lacustris (Figueiredo et al., 2021). An analysis using mitochondrial (mtDNA) markers was unable to differentiate between G. lacustris and G. labiatus from ETM, although both were distinct from G. labiatus from ELP. The authors also observed extensive haplotype sharing within the ETM and dated the separation between G. labiatus and G. lacustris in the ETM as no older than the last glacial maximum, while the separation between G. labiatus ETM and ELP populations dated back to the Pliocene (~3.6 mya). Morphological data showed significant differences between G. labiatus, G. labiatus ETM and G. labiatus ELP, although no feature was able to differentiate these three lineages simultaneously. Two potentially adaptively relevant traits showed important differences among groups. Lip size was significantly larger for G. labiatus (in both ELP and ETM), while color pattern was significantly different between ELP and ETM (both G. labiatus and G. lacustris). However, lip size variation within ELP (where only G. labiatus occurs) was comparable to that seen in ETM as a whole (where, traditionally, both species occur). This study concluded that ecoregions reflect the evolutionary history of these lineages more accurately, and that the ETM populations should be considered a single taxonomic species, due to the importance of color pattern in sexual selection and subsequent reproductive isolation, while differences in lip morphology would be better understood as phenotypic plasticity (Fig. 1; Figueiredo et al., 2021).

FIGURE 1| Map with the distribution of the individuals used in the present study. The hydrographic basins corresponding to ELP and ETM are indicated in different shades of gray. The gray lines represent the watershed between basins or sub-basins in the study area. Sample sites for Gymnogeophagus labiatus are shown in orange (TA – triangles; DC – squares; SC – circles. Sample sites for G. lacustris are shown in blue (ETMR – squares; ETML – circles) Species’ distributions according to Figueiredo et al. (2021).

However, mtDNA has a limited power in distinguishing independent genealogical units. First, because of its non-recombining nature, mtDNA represents a single locus, such that its specific genealogical structure may be due to chance alone (i.e., random genetic drift), introgression, or other evolutionary processes that are not representative of the genealogical structure of these terminals at the genome level (Boussau, Scornavacca, 2020). Second, Figueiredo et al. (2021) have shown that riverine and lacustrine populations in ETM have significant differences in lip morphology. Even if they interpret mtDNA data as suggesting that all individuals from ETM are G. lacustris, an alternative hypothesis is that riverine and lacustrine populations constitute different genealogical units (species), but that mtDNA did not have enough resolution to discriminate between them because the lake system in ETM is very recent compared to mtDNA mutation rate (Hickerson et al., 2006).

In the present study, we use a much larger, genome-wide data to test the speciation scenario proposed by Figueiredo et al. (2021) and to investigate the relationship between the riverine (thick lip) and lacustrine (thin lip) ecomorphs in ETM. More specifically, we ask first, do ETM and ELP represent different evolutionary lineages (corresponding to G. lacustris and G. labiatus, respectively, as defined by Figueiredo et al.)? Second, do riverine and lacustrine ecomorphs in ETM correspond to distinct genealogical units? Third, does genome-wide data support the notion that lip-size variation only reflects phenotypic plasticity?

Material and methods

---

Sampling and genomic data. All individuals used in the study were deposited in the scientific collection of the Laboratory of Ichthyology, Departamento de Zoologia, Universidade Federal do Rio Grande do Sul, Porto Alegre (UFRGS). Taxonomic identification of all specimens was based on their external morphology (Reis, Malabarba, 1988) and mitochondrial data (Figueiredo et al., 2021). In other words, all individuals from ETM were considered G. lacustris, and all individuals from ELP were considered G. labiatus. These ecoregions thus represent the two major units we tested for genome-wide genealogical distinctiveness. Within ELP, we defined distinct (genetic) populations based on the results of the genetic structure analyses (Tab. S1), which were then used in downstream analyses. Within ETM, for which the overall genetic structure was much lower (see Results), we considered individuals coming from riverine (ETMR) and lacustrine (ETML) environments as distinct (ecological) populations, as this was also consistent with the morphological difference in lip size between them (Tab. S1). Overall, we used 60 individuals (44 G. lacustris and 16 G. labiatus) for the genomic analysis (Fig. 1; Tabs. 1, S1).

TABLE 1 | Individuals and locations of Gymnogeophagus labiatus and G. lacustris included in the present study. ¹Population labels in Ecorregião Tramandaí-Mampituba (ETM) correspond to ecological populations (i.e., riverine (ETMR) or lacustrine (ETML) sites), while in Ecorregião Laguna dos Patos (ELP), population labels indicate genetic populations as defined in genetic structure analyses. ²Raw sequence data is provided for all individuals as SRA, but only those passing the filters for coverage/missing were included in the datasets analyzed in this study. *These individuals have originally been recognized as G. labiatus according to Reis, Malabarba (1988), but see Figueiredo et al. (2021) and the present study for a discussion of their taxonomic status. Populations: DC = Depressão Central, SC = Sinos-Caí, TA = Taquari-Antas.

Species	Ecoregion	Population1	Drainage	Localities	Sample size2
G. lacustris	ETM	ETML	Lagoa Bacopari	1	4	2
G. lacustris	ETM	ETML	Lagoa Corvina	1	7	1
G. lacustris	ETM	ETML	Lagoa dos Quadros	3	3	2
G. lacustris	ETM	ETML	Lagoa da Fortaleza	1	1	1
G. lacustris	ETM	ETML	Lagoa Emboaba	1	4	2
G. lacustris	ETM	ETML	Lagoa Emboabinha	1	1	1
G. lacustris	ETM	ETML	Lagoa da Pinguela	2	6	4
G. lacustris*	ETM	ETMR	Rio Maquiné	6	13	10
G. lacustris*	ETM	ETMR	Rio Três Forquilhas	1	2	1
G. lacustris*	ETM	ETMR	Rio Mampituba	2	3	1
G. labiatus	ELP	DC	Bacia do baixo rio Jacuí	2	3	2
G. labiatus	ELP	DC	Bacia do alto rio Jacuí	1	1	1
G. labiatus	ELP	DC	Bacia do lago Guaíba	2	3	3
G. labiatus	ELP	TA	Bacia do rio Taquari-Antas	3	5	5
G. labiatus	ELP	SC	Bacia do rio dos Sinos	1	2	2
G. labiatus	ELP	SC	Bacia do rio Caí	1	2	2

ELP includes all the interconnected rivers and lagoons that drain into the laguna dos Patos estuary. ETM includes all the interconnected lakes that ultimately drain into rio Tramandaí and can be further divided into two subsystems of interconnected lakes (Schwarzbold, Schäfer, 1984): one north of laguna Tramandaí (formed by rio Tramandaí itself, lagoa Itapeva, lagoa dos Quadros, and a set of small lakes in the municipality of Osório), and one to the south (including a series of interconnected lakes that reach lagoa da Porteira). We also included individuals sampled from isolated coastal lakes (lagoa Bacopari and lagoa Corvina) that are currently mapped to ELP (Abell et al., 2008) to understand the origin of their fish fauna. Based on the results considering the whole sample, these individuals were later included in the ETM dataset (see Results).

For molecular analyses, genomic DNA was obtained using a method adapted from Doyle, Doyle (1987) based on cetyltrimethylammonium bromide (CTAB), and extractions were stored at -20°C. After verification of genomic DNA concentrations, DNA quality was quantified with the Qubit^TM 3 fluorometer. Subsequently, the libraries were constructed using only samples with concentrations ≥ 30ng/μl. Two RADSeq libraries were constructed following the double digestion protocol of Peterson et al. (2012). Each library contains 30 individuals of both species of Gymnogeophagus. Succinctly, DNA was double digested with two restriction enzymes (SphI and MluCl), followed by a PCR ligation and amplification step, where unique barcodes (5 bp) and Illumina adapters were added to the digested DNA so that individuals could be grouped. The PCR products were cleaned, and fragments between 200–300 bp were selected (target size: 250 bp) from the pooled DNA after agarose gel electrophoresis by slicing the gel and recovering the DNA using magnetic beads. Libraries were sequenced on an Illumina HiSeq 2500, targeting single-end 100 bp reads. Sequencing generated approximately 140 million total reads per library that passed initial quality control at the sequencing facility.

We used the same filtering steps to create three SNP datasets, one for ETM, one for ELP and a general dataset including individuals from both ecoregions. Initially, raw reads were de-multiplexed and filtered using the process_radtags.pl script in STACKS v. 2.64 (Catchen et al., 2011, 2013) using the default configurations, and reads with Phred < 33 were discarded. We also used STACKS to assemble putative loci from demultiplexed reads. Next, we used USTACKS to reassemble the reads from each individual, which allows the detection of SNPs in each locus. We used a minimum depth coverage of three to create a stack (-m 3), allowing a maximum distance of two nucleotides among stacks (-M 2). The Removal (-r) and Deleveraging (-d) algorithms, were used to remove overrepresented stacks and resolve over merged loci, respectively, with the equal model type bounded (–model_type), and an error bounded for ε of 0.1 (–bound_high) (Catchen et al., 2013). From the sample processed by USTACKS a catalog of genomic sequences (i.e., consensus homozygous loci) was constructed in CSTACKS, creating a set of consensus loci by merging the alleles. Finally, the loci for all individuals were matched to the assembled contigs using SSTACKS by searching the set of stacks (constructed by USTACKS) against the catalog (created by CSTACKS). The search was run under default options, allowing two mismatches between individuals (-n 2).

From SSTACKS the POPULATIONS module was run for the three datasets under loose parameters (-r 0 -p 2 -m 5 –min_maf 0 –max_obs_het 0.5), creating a .vcf output with all SNPs per locus. Loci in the upper 95% quantile of the genetic diversity (theta) distribution were flagged as either potential sequencing artefacts or assembly errors and removed (Thomaz et al., 2017). In addition, the number of SNPs per sequence position was plotted and the 10 positions at the 3’ end with an inflated number of SNPs (compared to other positions) were manually trimmed. A whitelist was created identifying the selected SNPs, and the POPULATIONS program was run again, exporting the new filtered vcf file.

The program PLINK 2 (Purcell et al., 2007) was then used to test the effect of missing data on the number of retained SNPs for each dataset. The full dataset was checked for missingness per SNP (–geno) and individual (–mind). Among the tested levels of missing data, all other analyses were performed with a dataset containing a maximum of 25% missing data per unlinked SNP and 10% per individual. These parameters were chosen to minimize the number of excluded individuals and maximize the number of SNPs retained. A final run of the POPULATIONS module was executed blacklisting loci with exceeding values for missing data. The –write-single-snp flag was used to take only one SNP per loci to avoid retaining loci in obvious linkage disequilibrium (LD). We generated three vcf files: one with all individuals (I), from which the other two datasets were derived, with individuals from ETM (II) or ELP (III) exclusively. The vcf file containing all retained SNPs can be found in https://10.5281/zenodo.11477978. Raw sequencing data is available from NCBI as Sequence Read Archive (SRA) associated with BioSamples SAMN44088444 – SAMN44088503.

Genetic structure and phylogenetic relationships. We performed a principal component analysis (PCA) to evaluate the general genetic structure in the data. The analysis was made for all datasets using the plink2 software, where the eigenvalues files were generated. The PCA was plotted using ggplot2 package in R v. 4.3.1 (R Development Core Team, 2023). For each dataset, the most likely number of genetic clusters (i.e., K-value) was estimated in ADMIXTURE (Alexander et al., 2009) which uses large autosomal SNP genotype datasets to group individuals into genetic populations and provides, for each individual, an estimate of its ancestry coming from each of the K genetic clusters. We explored K-values ranging from one (a single panmictic population) to N, where N represents the number of populations in each dataset. The best K-value was selected based on a cross-validation procedure (–cv), where the best value of K had the lowest cross-validation error value.

We used SVDQUARTETS (Chifman, Kubatko, 2014) to infer the genealogical relationships among populations based on a coalescent population model. The NEXUS file containing the more inclusive SNP dataset was run in PAUP* v. 4.0a166 (Swofford, 2003). The tree resulting from the SVDQUARTETS analysis was rooted according to the mtDNA tree (Figueiredo et al., 2021), assuming that G. labiatus and G. lacustris represent reciprocally monophyletic groups. The occurrence of gene flow among populations, which is indicated by negative and significant values for the f3-statistic, was tested in the ADMIXTOOLS program (Maier, Patterson, 2024). Finally, basic genetic diversity statistics (H_O, H_E and F_IS), as well as pairwise F_ST values between populations were estimated with the DartT package (Gruber et al., 2018) in R. Genetic diversity analyses considered rivers and lakes in ETM as distinct populations (ETMR and ETML, respectively), as these were considered different species until recently. In ELP, we used the results from genetic structure analysis to identify three populations: Taquari-Antas (TA), including samples from drainages associated with the the Taquari-Antas drainage, Sinos-Caí (SC), including samples associated with drainages of rio dos Sinos and rio Caí, Depressão Central (DC), including samples from the upper and lower rio Jacuí system, and lago Guaíba.

Results​

---

Three basic data sets were created from the STACKS assemblies. The complete dataset (Dataset I) consisted of 41 individuals and 29,747 variant sites (SNPs). This dataset was used in all subsequent analyses including both ETM and ELP lineages (species). Individuals from ETM and isolated coastal lagoons (see Material and Methods) were included in Dataset II, which consisted of 26 individuals and 16,920 SNPs. Finally, the 15 individuals from ELP were included in Dataset III, containing 26,478 SNPs. These two latter datasets aimed at maximizing the power to detect genetic structure within each drainage (species), considering missing data levels for the relevant individuals in each case.

The PCA for Dataset I showed a clear separation between ETM and ELP individuals associated to PC1 (that contained 35.6% of the total variance), while PC2 (9.98% of the total variance) separated individuals from Taquari-Antas (TA) from other ELP drainages (Fig. 2A). For Dataset II, an initial analysis showed the individual from rio Mampituba very separated from the other samples in PC1 (Fig. S2) When this sample was excluded, PC1 and PC2 retained only 5.93% and 5.5% of the total variance, respectively, and showed no clear separation between ETMR and ETML, which would be expected based on the traditional taxonomic classification that assigned them to different species (Fig. 2B). Finally, Dataset III revealed a separation of individuals from TA associated with PC1 (18.4% of the total variation), while PC2 (10.4% of the total variation) separated individuals from SC apart from those coming from DC) (Fig. 2C).

FIGURE 2| PCA for all datasets. Individuals from Gymnogeophagus labiatus are shown in orange, while those from G. lacustris are shown in blue. A. Dataset I, including both species; B. Dataset II, including only G. lacustris; C. Dataset III, including only G. labiatus. See the meanings of the codes in Tab. 1.

ADMIXTURE identified K = 2 as the best K value for Dataset I. However, the difference in cross-validation error between K = 2 and K = 3 was very small (0.335 and 0.359, respectively). While for K = 2 there was a separation between ecoregions (i.e., ETM vs. ELP), with K = 3 TA formed a distinct genetic cluster within ELP (Fig. 3A). The consideration of higher K-values resulted in additional genetic components associated to each river in ELP, consistent with a lack of further structure within ETM (Figs. S3, S4, S5). The SVDQUARTETS analysis of all 29,747 SNPs resulted in a fully resolved species tree in which ETM and ELP form reciprocally monophyletic groups with maximum support. Internal relationships were also fully resolved, showing, in ELP, a sister relationship between DC and SC regarding TA (Fig. 3B).

FIGURE 3| A. Admixture analysis for Dataset I considering K = 2 (above) and K = 3 (below). Each column corresponds to the different individuals. The fraction of genetic ancestry in each of the K genetic clusters is represented in different colors. B. SVDQUARTETS tree based on a population coalescent model. Populations from the same ecoregions are highlighted in different colors. See the meanings of the codes in Tab. 1.

Genetic diversity was highest in SC, followed by DC, TA, ETMR and then ETML (Tab. 2). While F_IS values seem a bit larger in TA compared to other drainages, this could result from differences in genetic structure within each drainage rather than differences in inbreeding levels. As expected, pairwise F_ST values were higher among populations between ELP vs. ETM (which represent different species), ranging between 0.699 (TA x ETMR) and 0.730 (SC x ETML). Within each ecoregion, ELP had higher structure, with F_ST values ranging between 0.114 (TA x DC) and 0.287 (SC vs. DC), while ETM had the lowest value (0.002, ETML x ETMR) (Tab. 3). The ADMIXTOOLS f3-statistic indicated, with strong statistical support, secondary contact between ETMR and the other populations (Tab. 4).

TABLE 2 | Average observed (H_O), expected (H_E) heterozygosity, and the corresponding inbreeding coefficient (F_IS) for each population. See the meanings of the codes in Tab. 1.

Lineage	ETM		ELP
	ETML	ETMR	DC	SC	TA
HO	0.054	0.055	0.109	0.120	0.086
HE	0.055	0.058	0.111	0.117	0.101
FIS	0.061	0.09	0.112	0.110	0.246

TABLE 3 | Pairwise F_ST values. P < 0.01 for all values. See the meanings of the codes in Tab. 1.

Lineage	ETM		ELP
	ETML	ETMR	DC	SC	TA
ETML	–	–	–	–	–
ETMR	0.002	–	–	–	–
DC	0.706	0.701	–	–	–
SC	0.730	0.724	0.287	–	–
TA	0.705	0.699	0.114	0.265	–

TABLE 4 | Test of population admixture based on the f3-statistic. Negative and significant z-scores indicate admixture between Pop1 and Pop2. P-values for positive z-scores were omitted, as they only indicate the amount of genetic drift for each population rather than admixture. See the meanings of the codes in Tab. 1.

Pop1	Pop2	Pop3	f3	SE	z-score	P-value
ETMR	DC	ETML	-0.000769	0.000133	-5.78	7.27E-09
ETMR	SC	ETML	-0.00076	0.000133	-5.73	1.02E-08
ETMR	TA	ETML	-0.000702	0.000133	-5.27	1.33E-07
ETML	ETMR	TA	0.00084	0.000138	6.08	–
ETML	ETMR	SC	0.000898	0.000137	6.55	–
ETML	ETMR	DC	0.000906	0.000138	6.59	–
SC	DC	ETML	0.00549	0.000496	11.1	–
SC	DC	ETMR	0.0055	0.000495	11.1	–
SC	TA	DC	0.00873	0.000436	20	–
DC	SC	TA	0.0106	0.00042	25.3	–
DC	ETML	SC	0.0138	0.000538	25.7	–
DC	ETMR	SC	0.0138	0.000537	25.8	–
SC	TA	ETMR	0.0165	0.000691	23.9	–
SC	TA	ETML	0.0166	0.000691	24	–
DC	ETMR	TA	0.0216	0.000718	30.2	–
DC	ETML	TA	0.0217	0.000718	30.2	–
TA	DC	ETML	0.0315	0.00085	37.1	–
TA	DC	ETMR	0.0316	0.00085	37.2	–
TA	ETML	SC	0.0348	0.000868	40	–
TA	ETMR	SC	0.0348	0.000867	40.1	–
TA	DC	SC	0.0426	0.000809	52.7	–
ETMR	SC	TA	0.156	0.00183	85.6	–
ETML	SC	TA	0.158	0.00184	86	–
ETMR	TA	DC	0.16	0.00184	86.6	–
ETML	TA	DC	0.161	0.00185	87	–
ETMR	SC	DC	0.167	0.00186	89.8	–
ETML	SC	DC	0.169	0.00188	90.2	–
SC	ETML	ETMR	0.174	0.0019	91.2	–
DC	ETML	ETMR	0.182	0.00194	93.9	–
TA	ETML	ETMR	0.192	0.00198	96.9	–

Discussion​

---

If we take an evolutionary perspective to answer the question “What is a species?”, we could use the Unified Species Concept and state that species are independent evolutionary lineages (Wiley, 1978; De Queiroz, 2007). The key issue, therefore, is how do we recognize which evolutionary lineages are independent enough to be considered true “species”, and why do some data sets but not others support this distinction (Hey, 2001; De Queiroz, 2007). For genetic data, it has become relatively common to use “objective” methods that integrate a set of gene trees to compare, based on some probabilistic method, how alternative classification schemes fit multi-species genealogical models (e.g., Yang, 2015; Jones, 2017). On its turn, morphological differentiation shows another layer in the process of lineage divergence, which may be faster or slower than genealogical divergence, allowing its use as an independent source of evidence to make taxonomic decisions about a given evolutionary process (Carew et al., 2005; Yeates et al., 2011).

In this study, we corroborate the findings of Figueiredo et al. (2021) based on mtDNA and morphological data, and show that Gymnogeophagus labiatus, as originally defined in Reis, Malabarba (1988), represent a set of paraphyletic lineages in regard to G. lacustris. Consequently, from a coalescent point of view, genome-wide SNP data indicate that the two ETM morphotypes (with or without hypertrophied lips, previously considered as representative of G. labiatus and G. lacustris, respectively) constitute a single evolutionary lineage: G. lacustris, while the populations in ELP correspond to G. labiatus. This is not unexpected, as there are other sister species pairs that follow the ETM/ELP divide, such as Diapoma itaimbe (Malabarba & Weitzman, 2003) vs. D. dicropotamicus (Malabarba & Weitzman, 2003) (Ito et al., 2022) and Mimagoniates rheocharis Menezes & Weitzman, 1990 vs. M. inequalis (Eigenmann, 1911) (Thomaz et al., 2015). We also found a very distinct pattern of genetic diversity and structure between these two lineages. While G. lacustris has virtually no internal genetic structure and low diversity (but see below), G. labiatus is both more diverse and more structured, especially considering the Taquari-Antas population.

Figueiredo et al. (2021) revealed significant morphological differences among G. labiatus and the two ecomorphs of G. lacustris, although no measurement was sufficient to discriminate among the three groups simultaneously. As expected from their original taxonomic description (Reis, Malabarba, 1988), lip area was the main trait differentiating G. labiatus and the riverine ecomorph of G. lacustris from the lacustrine ecomorph. On the other hand, other measurements that showed significant morphological variation (pectoral fin size and snout size) discriminated between G. labiatus and G. lacustris. The analysis of live coloration further corroborated the differentiation according to river basins, the most striking differences being lip color, which was conspicuously reddish in individuals from G. lacustris, and the bright blue spots on the cheeks, which are more numerous in individuals from G. labiatus. The differences in coloration between river basins agree with the genomic data that indicates that distinct evolutionary units in each ecoregion. It has already been shown that color diversity among cichlids, especially in male nuptial coloration, is the result of strong sexual and natural selection (Seehausen et al., 1999; Kocher, 2004; Barson et al., 2007). Sexual selection associated with mate choice has a clear relationship with speciation, which has led many to propose that sexual selection on nuptial coloration is probably one of the main factors driving the diversification of cichlids (Kocher, 2004; Wagner et al., 2012) given its importance as a reproductive isolation mechanism.

The results for lip area are at odds with those obtained for other morphological measurements and genetic (genomic and mtDNA) markers, which could call into question its usefulness in delimiting taxonomic groups. Hypertrophied lips are a rare phenotype, but they have arisen independently in all the main cichlid clades (Burress, 2015). The development of hypertrophied lips may or may not be associated with a change in diet from the ancestral state, but it is consistently linked to specialization in foraging oriented towards rocky crevices (Baumgarten et al., 2015). In addition, hypertrophied lips are also a polymorphism often associated with incipient species pairs (Elmer et al., 2010; Colombo et al., 2013; Manousaki et al., 2013; Machado-Schiaffino et al., 2014). Its trophic function seems to be associated with foraging in rocky crevices (Baumgarten et al., 2015). The wide variety of habitats found in each river basin could explain the variation in lip size found within and between populations. This could be because both basins are made up of stony-bottomed rivers with lotic waters and strong currents, and sandy-bottomed lakes and lagoons with lentic waters. Thus, variation in lip size could be attributed to phenotypic plasticity, and individuals with hypertrophied lips could be considered as ecomorphs within their respective species (Figueiredo et al., 2021). Phenotypic plasticity of lip size was reported in Amphilophus labiatus (Günther, 1864) from Lake Nicaragua, possibly playing an adaptive role in foraging (Machado-Schiaffino et al., 2014). Alternatively, the different ecomorphs of G. lacustris could indicate incipient speciation, in which (adaptive) morphological divergence is faster than (neutral) genealogical divergence. Under this interpretation, these ecomorphs could be in the speciation “gray zone” (De Queiroz, 2007; Roux et al., 2016). Recent speciation between riverine and lacustrine habitats in the Tramandaí-Mampituba ecoregion has been recently described for Microglanis lucenai Lehmann, Bartzen & Malabarba, 2024 and M. cibelae Malabarba & Mahler, 1998 (Lehmann et al., 2024).

Unexpectedly, we found evidence for a secondary contact between G. lacustris riverine (large-lipped) ecomorph) and all populations of G. labiatus. This suggests that large lips in these species may not represent convergence or phenotypic plasticity. Rather, they indicate a shared genetic background in both species, representing a possible case of adaptive introgression (Arnold, Kunte, 2017). The evolutionary history of cichlids is full of admixture and introgression events, many of which have been associated with adaptive radiations (Meier et al., 2017; Svardal et al., 2020; Singh et al., 2022). Interestingly, genetic analysis based on RAD-Seq data suggests that the large lips in Amphilophus labiatus from Lakes Nicaragua and Managua also spread via adaptive introgression (Sowersby et al., 2021). Given that all G. labiatus populations show evidence of contributing to the riverine ecomorph of G. lacustris, a possible scenario is that the secondary contact occurred after the divergence between G. labiatus nad G. lacustris, but before the divergence among G. labiatus populations, which, based on the mtDNA scenario proposed by Figueiredo et al. (2021) correspond to about ~3.6 mya (95% CI 1.1 mya – 6.0 mya). Genetic introgression does not invalidate the finding that individuals in different ecoregions correspond to distinct species (i.e., independent evolutionary lineages) but offers an alternative explanation for the hypertrophied lips shown by the riverine individuals of G. lacustris.

Concerning G. lacustris, further studies will be necessary to better understand the level of genetic structure in rio Mampituba populations, and the pattern of gene flow among riverine and lacustrine populations along the ecological gradient river/lagoon. Another fundamental question is whether the signal of introgression between ELP and ETMR is associated to lip morphology. Regarding G. labiatus, future work should focus on increasing geographic sampling to include the tributaries of laguna dos Patos south of rio Jacuí, and the tributaries of laguna Mirim. This would be important to understand the evolutionary meaning of the deeply divergent mtDNA lineages in rio Camaquã (a tributary of laguna dos Patos) identified by Figueiredo et al. (2021). In particular, the area around these lagunas shows a remarkable diversity of annual killifishes (e.g., Garcez et al., 2020; Volcan et al., 2021), whose diversification may have started as early as ~2.7 mya (Fernandes et al., 2021). However, while killifish ecology favors quick diversification (Garcez et al., 2020), G. labiatus need proper riverine connections to, first, originate new populations, and, second, maintain gene flow among them. Differently from the northern part of its distribution, where different subdrainages are ultimately connected to lago Guaíba, southern drainages flew in isolation to the Atlantic Ocean until the emergence of laguna Mirim and laguna dos Patos (around 200,000 and 120,000 years ago, respectively) (Tomazelli, Villwock, 2000; Weschenfelder et al., 2010). In this case, recent divergence times among southern drainages would reflect the effect of the hydrographic connections originated during the Middle/Late Pleistocene, while older divergence times might reflect colonization from river capture events.

Acknowledgments​

---

PICCF received a PhD scholarship from CAPES. ALSV and MV receive PhD scholarships from CNPq. LRM is supported by CNPq, grants number: 307890/2016–3 and 401204/2016–2. NJRF is supported by CNPq, grant number: 316900/2023–0 and FAPERGS, grant number: 19/2551–0001929–7. We are also thankful to Hernán López-Fernández, Bruno Melo and two anonymous reviewers for their helpful comments in the first version of the manuscript.

References​

---

Abell R, Thieme ML, Revenga C, Bryer M, Kottelat M, Bogutskaya N et al. Freshwater ecoregions of the World: a new map of biogeographic units for freshwater biodiversity conservation. Bioscience. 2008; 58(5):403–14. https://doi.org/10.1641/B580507

Alexander DH, Novembre J, Lange K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 2009; 19(9):1655–64. https://doi.org/10.1101/gr.094052.109

Alonso F, Terán GE, Aguilera G, Říčan O, Casciotta J, Serra WS et al. Description of a new species of the Neotropical cichlid genus Gymnogeophagus Miranda Ribeiro, 1918 (Teleostei: Cichliformes) from the Middle Paraná basin, Misiones, Argentina. PLoS ONE. 2019; 14(2):e0210166. https://doi.org/10.1371/journal.pone.0210166

Arnold ML, Kunte K. Adaptive genetic exchange: a tangled history of admixture and evolutionary innovation. Trends Ecol Evol. 2017; 32(8):601–11. https://doi.org/10.1016/j.tree.2017.05.007

Barson NJ, Knight ME, Turner GF. The genetic architecture of male colour differences between a sympatric Lake Malawi cichlid species pair. J Evol Biol. 2007; 20(1):45–53. https://doi.org/10.1111/j.1420-9101.2006.01228.x

Baumgarten L, Machado-Schiaffino G, Henning F, Meyer A. What big lips are good for: on the adaptive function of repeatedly evolved hypertrophied lips of cichlid fishes. Biol J Linn Soc. 2015; 115(2):448–55. https://doi.org/10.1111/bij.12502

Becker FG, Fries LCC, Ferrer J, Bertaco VA, Luz-Agostinho KDG, Silva JP, et al. Fishes of the Taquari-Antas river basin (Patos Lagoon basin), southern Brazil. Brazilian J Biol. 2013; 73(1):79–90. https://doi.org/10.1590/S1519-69842013000100010

Boussau B, Scornavacca C. Reconciling gene trees with species trees. In: Scornavacca C, Delsuc F, Galtier N, editors. Phylogenetics in the Genomic Era. Montpellier: HAL; 2020. p.1–23. Available from: https://hal.inria.fr/PGE/hal-02535529

Burress ED. Cichlid fishes as models of ecological diversification: patterns, mechanisms, and consequences. Hydrobiologia. 2015; 748(1):7–27. https://doi.org/10.1007/s10750-014-1960-z

Carew ME, Pettigrove V, Hoffmann AA. The utility of dna markers in classical taxonomy: using cytochrome oxidase i markers to differentiate Australian Cladopelma (Diptera: Chironomidae) midges. Ann Entomol Soc Am. 2005; 98(4):587–94. https://doi.org/10.1603/0013-8746(2005)098[0587:TUODMI]2.0.CO;2

Casciotta J, Almirón A, Piálek L, Říčan O. Gymnogeophagus taroba (Teleostei: Cichlidae), a new species from the río Iguazú basin, Misiones, Argentina. História Natural, Tercera Serie. 2017; 7(2):5–22.

Catchen JM, Amores A, Hohenlohe P, Cresko W, Postlethwait JH. Stacks: building and genotyping loci de novo from short-read sequences. G3. 2011; 1(3):171–82. https://doi.org/10.1534/g3.111.000240

Catchen, J, Hohenlohe PA, Bassham S, Amores A, Cresko WA. Stacks: an analysis tool set for population genomics. Mol Ecol. 2013; 22(11):3124–40. https://doi.org/10.1111/mec.12354

Chifman, J, Kubatko L. Quartet inference from SNP data under the coalescent model. Bioinformatics. 2014; 30(23):3317–24. https://doi.org/10.1093/bioinformatics/btu530

Colombo M, Diepeveen ET, Muschick M, Santos ME, Indermaur A, Boileau N et al. The ecological and genetic basis of convergent thick-lipped phenotypes in cichlid fishes. Mol Ecol. 2013; 22(3):670–84. https://doi.org/10.1111/mec.12029

De Queiroz K. Species concepts and species delimitation. Syst Biol. 2007; 56(6):879–86. https://doi.org/10.1080/10635150701701083

Doyle JJ, Doyle JL. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin. 1996; 19(1):11–15.

Elmer KR, Lehtonen TK, Kautt AF, Harrod C, Meyer A. Rapid sympatric ecological differentiation of crater lake cichlid fishes within historic times. BMC Biol. 2010; 8:60. https://doi.org/10.1186/1741-7007-8-60

Fernandes MO, Barbosa C, Garcez DK, Varela Junior AS, Volcan MV, Robe LJ. Phylogeographic analyses and taxonomic inconsistencies of the Neotropical annual fish Austrolebias minuano, Austrolebias charrua and Austrolebias pongondo (Cyprinodontiformes: Rivulidae). Environ Biol Fish. 2021; 104:1–14. https://doi.org/10.1007/s10641-020-01045-9

Figueiredo PICC, Malabarba LR, Fagundes NJR. Hydrography rather than lip morphology better explains the evolutionary relationship between Gymnogeophagus labiatus and G. lacustris in Southern Brazil (Cichlidae: Geophagini). Neotrop Ichthyol. 2021; 19(4):e200154. https://doi.org/10.1590/1982-0224-2020-0154

Fricke R, Eschmeyer WN, Van der Laan R. Eschmeyer’s catalog of fishes: genera, species, references [Internet]. San Francisco: California Academy of Science; 2024. Available from: http://researcharchive.calacademy.org/research/ichthyology/catalog/fishcatmain.asp

Garcez DK, Fernandes MO, Ozório GR, Volcan MV, Robe LJ. Phylogenetic structure of Neotropical annual fish of the genus Cynopoecilus (Cyprinodontiformes: Rivulidae), with an assessment of taxonomic implications. J Zool Syst Evol Res. 2020; 58:1123–34. https://doi.org/10.1111/jzs.12389

Gruber B, Unmack PJ, Berry OF, Georges A. Dartr: an r package to facilitate analysis of SNP data generated from reduced representation genome sequencing. Mol Ecol Res. 2018; 18(3):691–99. https://doi.org/10.1111/1755-0998.12745

Henning F, Meyer A. The evolutionary genomics of cichlid fishes: explosive speciation and adaptation in the postgenomic era. Annu Rev Genomics Hum Genet. 2014; 15(1):417–41. https://doi.org/10.1146/annurev-genom-090413-025412

Hey J. The mind of the species problem. Trends Ecol Evol. 2001; 16(7):326–29. https://doi.org/10.1016/S0169-5347(01)02145-0

Hickerson MJ, Meyer CP, Moritz C. DNA barcoding will often fail to discover new animal species over broad parameter space. Syst Biol. 2006; 55(5):729–39. https://doi.org/10.1080/10635150600969898

Ito PMM, Carvalho TP, Pavanelli CS, Vanegas-Ríos JA, Malabarba LR. Phylogenetic relationships and description of two new species of Diapoma (Characidae: Stevardiinae) from the La Plata River basin. Neotrop Ichthyol. 2022; 20(1):e210115. https://doi.org/10.1590/1982-0224-2021-0115

Jones G. Algorithmic improvements to species delimitation and phylogeny estimation under the multispecies coalescent. J Math Biol. 2017; 74(1–2):447–67. https://doi.org/10.1007/s00285-016-1034-0

Kocher TD. Adaptive evolution and explosive speciation: the cichlid fish model. Nat Rev Genet. 2004; 5(4):288–98. https://doi.org/10.1038/nrg1316

Langeani F, Buckup PA, Malabarba LR, Py-Daniel LH, Lucena CAS, Rosa RS et al. Peixes de água doce. In: Estado da arte e perspectivas para a zoologia no Brasil. Curitiba: Sociedade Brasileira de Zoologia; Ed. UFPr; 2009. p.211–30.

Lehmann A. P, Bartzen CS, Malabarba LR. The barcode trap — description of a new species of Microglanis, with a review of the status of Microglanis cibelae (Siluriformes: Pseudopimelodidae). J Fish Biol. 2024; 105(1):110–23. https://doi.org/10.1111/jfb.15764

López-Fernández H, Winemiller KO, Honeycutt RL. Multilocus phylogeny and rapid radiations in Neotropical cichlid fishes (Perciformes: Cichlidae: Cichlinae). Mol Phylogenet Evol. 2010; 55(3):1070–86. https://doi.org/10.1016/j.ympev.2010.02.020

Loureiro M, Zarucki M, Malabarba LR, González-Bergonzoni I. A new species of Gymnogeophagus Miranda Ribeiro from Uruguay (Teleostei: Cichliformes). Neotrop Ichthyol. 2016; 14(1):e150082. https://doi.org/10.1590/1982-0224-20150082

Lowe-McConnell RH. Estudos ecológicos de comunidades de peixes tropicais. São Paulo: EDUSP; 1999.

Machado-Schiaffino G, Henning F, Meyer A. Species-specific differences in adaptive phenotypic plasticity in an ecologically relevant trophic trait: hypertrophic lips in midas cichlid fishes. Evolution. 2014; 68(7):2086–91. https://doi.org/10.1111/evo.12367

Maier R, Patterson N. admixtools: inferring demographic history from genetic data. R package version 2.0.4. 2024. Available from: https://github.com/uqrmaie1/admixtools

Malabarba LR, Isaia EA. 1992. The fresh-water fish fauna of the Rio Tramandaí Drainage, Rio Grande do Sul, Brazil, with a discussion of its historical origin. Comun Mus Ciênc Tecnol PUCRS, sér zool. 5(12):97–223.

Malabarba LR, Malabarba MC, Reis RE. Descriptions of five new species of the Neotropical cichlid genus Gymnogeophagus Miranda Ribeiro, 1918 (Teleostei: Cichliformes) from the rio Uruguay drainage. Neotrop Ichthyol. 2015; 13(4):637–62. https://doi.org/10.1590/1982-0224-20140188

Malabarba LR, Neto PC, Bertaco VA, Carvalho TP, Santos JF, Artioli LGS. Guia de identificação dos peixes da bacia do rio Tramandaí. Porto Alegre: Via Sapiens; 2013.

Manousaki T, Hull PM, Kusche H, MacHado-Schiaffino G, Franchini P, Harrod C et al. Parsing parallel evolution: ecological divergence and differential gene expression in the adaptive radiations of thick-lipped Midas cichlid fishes from Nicaragua. Mol Ecol. 2013; 22(3):650–69. https://doi.org/10.1111/mec.12034

Masonick P, Meyer A, Hulsey CD. Phylogenomic analyses show repeated evolution of hypertrophied lips among Lake Malawi cichlid fishes. Genome Biol Evol. 2022; 14(4):evac051. https://doi.org/10.1093/gbe/evac051

Meier JI, Marques DA, Mwaiko S, Wagner CE, Excoffier L, Seehausen O. Ancient hybridization fuels rapid cichlid fish adaptive radiations. Nat Commun. 2017; 8:14363. https://doi.org/10.1038/ncomms14363

Peterson BK, Weber JN, Kay EH, Fisher HS, Hoekstra HE. Double digest RADseq: an inexpensive method for de novo SNP discovery and genotyping in model and non-model species. PLoS ONE. 2012; 7(5):e37135. https://doi.org/10.1371/journal.pone.0037135

Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D et al. PLINK: a toolset for whole-genome association and population-based linkage analysis. Am J Hum Genet. 2007; 81(3):559–75. https://doi.org/10.1086/519795

R Development Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2023. Available from: https://www.r-project.org/

Reis RE, Malabarba LR. Revision of the neotropical cichlid genus Gymnogeophagus Ribeiro, 1918, with descriptions of two new species (Pisces, Perciformes). Rev Bras Zool. 1988; 4(4):259–305. https://doi.org/10.1590/S0101-81751987000400002

Renne PR, Ernesto M, Pacca IG, Coe RS, Glen JM, Prévot M et al. The age of Paraná flood volcanism, rifting of Gondwanaland, and the Jurassic-Cretaceous boundary. Science. 1992; 258(5084):975–99. https://doi.org/10.1126/science.258.5084.975

Říčan O, Říčanová Š, Dragová K, Piálek L, Almirón A, Casciotta J. Species diversity in Gymnogeophagus (Teleostei: Cichlidae) and comparative biogeography of cichlids in the Middle Paraná basin, an emerging hotspot of fish endemism. Hydrobiologia. 2019; 832: 331–54. https://doi.org/10.1007/s10750-018-3691-z

Roux C, Fraïsse C, Romiguier J, Anciaux Y, Galtier N, Bierne N. Shedding light on the grey zone of speciation along a continuum of genomic divergence. PLoS Biol. 2016; 14(12):e2000234. https://doi.org/10.1371/journal.pbio.2000234

Schwarzbold A, Schäfer A. Gênese e morfologia das lagoas costeiras do Rio Grande do Sul – Brasil. Amazoniana. 1984; 9(1):87–104.

Seehausen O, Mayhew PJ, Van Alphen JJM. Evolution of colour patterns in East African cichlid fish. J Evol Biol. 1999; 12(3):514–34. https://doi.org/10.1046/j.1420-9101.1999.00055.x

Singh P, Irisarri I, Torres-Dowdall J, Thallinger GG, Svardal H, Lemmon EM et al. Phylogenomics of trophically diverse cichlids disentangles processes driving adaptive radiation and repeated trophic transitions. Ecol Evol. 2022; 12(7):e9077. https://doi.org/10.1002/ece3.9077

Sowersby W, Cerca J, Wong BBM, Lehtonen TK, Chapple DG, Leal-Cardín M et al. Pervasive admixture and the spread of a large-lipped form in a cichlid fish radiation. Mol Ecol. 2021; 30(21):5551–71. https://doi.org/10.1111/mec.16139

Svardal H, Quah FX, Malinsky M, Ngatunga BP, Miska EA, Salzburger W et al. Ancestral hybridization facilitated species diversification in the Lake Malawi cichlid fish adaptive radiation. Mol Biol Evol. 2020; 37(4):1100–13. https://doi.org/10.1093/molbev/msz294

Swofford DL. PAUP* Phylogenetic analysis using parsimony (* and other methods). Version 4. 2003. Available from: http://paup.csit.fsu.edu/

Thomaz AT, Arcila D, Ortí G, Malabarba LR. Molecular phylogeny of the subfamily Stevardiinae Gill, 1858 (Characiformes: Characidae): classification and the evolution of reproductive traits. BMC Evol Biol. 2015; 15:146. https://doi.org/10.1186/s12862-015-0403-4

Thomaz AT, Malabarba LR, Knowles LL. Genomic signatures of paleodrainages in a freshwater fish along the southeastern coast of Brazil: genetic structure reflects past riverine properties. Heredity. 2017; 119(4):287–94. https://doi.org/10.1038/hdy.2017.46

Tomazelli LJ, Villwock JA. O Cenozóico costeiro do Rio Grande do Sul. In: Holz M, Ros LF, editors. Geologia do Rio Grande do Sul. Porto Alegre: Edições CIGO/UFRGS; 2000. p.375–406.

Turcati A, Serra-Alanis WS, Malabarba LR. A new mouth brooder species of Gymnogeophagus with hypertrophied lips (Cichliformes: Cichlidae). Neotrop Ichthyol. 2018; 16(4):e180118. https://doi.org/10.1590/1982-0224-20180118

Urban S, Gerwin J, Hulsey CD, Meyer A, Kratochwil CF. The repeated evolution of stripe patterns is correlated with body morphology in the adaptive radiations of East African cichlid fishes. Ecol Evol. 2022; 12(2):e8568. https://doi.org/10.1002/ece3.8568

Villwock AJ. Geology of the coastal province of Rio Grande do Sul, Southern Brazil. A Synthesis. Pesquisas em Geociências. 1984; 16(16):5–49.

Volcan MV, Barbosa C, Robe LJ, Lanés LEK. Molecular phylogeny of the Austrolebias adloffi group (Cyprinodontiformes, Rivulidae), with description of two new endangered and highly endemic species of annual killifishes from the Laguna dos Patos system, southern Brazil. Zootaxa. 2021; 4965(1):87–113. http://doi.org/10.11646/zootaxa.4965.1.4

Wagner CE, Harmon LJ, Seehausen O. Ecological opportunity and sexual selection together predict adaptive radiation. Nature. 2012; 487(7407):366–69. https://doi.org/10.1038/nature11144

Weschenfelder J, Corrêa ICS, Aliotta S, Baitelli R. Paleochannels related to late quaternary sea-level changes in Southern Brazil. Braz J Oceanogr. 2010; 58:35–44.

Wiley EO. The evolutionary species concept reconsidered. Syst Zool. 1978; 27(1):17–26. https://doi.org/10.2307/2412809

Yang Z. A tutorial of BPP for species tree estimation and species delimitation. Curr Zool. 2015; 61(5):854–65. https://doi.org/10.1093/czoolo/61.5.854

Yeates DK, Seago A, Nelson L, Cameron SL, Joseph L, Trueman JWH. Integrative taxonomy, or iterative taxonomy? Syst Entomol. 2011; 36(2):209–17. https://doi.org/10.1111/j.1365-3113.2010.00558.x

Zachos FE. (New) Species concepts, species delimitation and the inherent limitations of taxonomy. J Genet. 2018; 97:811–15. https://doi.org/10.1007/s12041-018-0965-1

Authors

---

Pedro Ivo C. C. Figueiredo¹, Diogo T. Hashimoto², Raquel B. Ariede², André L. S. Zani¹, Maikel Varal¹, Luiz R. Malabarba^3,4 and Nelson J. R. Fagundes^1,3,5

^[1]    Programa de Pós-Graduação em Genética e Biologia Molecular, Instituto de Biociências, Universidade Federal do Rio Grande do Sul, 91501-970 Porto Alegre, RS, Brazil. (PICCF) pedrocampani@gmail.com, (ALSZ) zani.andre@yahoo.com.br, (MV) maikelvaral@yahoo.com.br, (NJRF) nelson.fagundes@ufrgs.br (corresponding author).

^[2]    Centro de Aquicultura da UNESP, Jaboticabal Campus, Universidade Estadual de São Paulo, 14884-900 Jaboticabal, SP, Brazil. (DTH) diogo.hashimoto@unesp.br, (RBA) raquel.ariede@gmail.com.

^[3]    Programa de Pós-Graduação em Biologia Animal, Instituto de Biociências, Universidade Federal do Rio Grande do Sul, 91501 970 Porto Alegre, RS, Brazil. (LRM) malabarb@ufrgs.br.

^[4]    Departamento de Zoologia, Instituto de Biociências, Universidade Federal do Rio Grande do Sul, 91501-970 Porto Alegre, RS, Brazil.

^[5]    Departamento de Genética, Instituto de Biociências, Universidade Federal do Rio Grande do Sul, 91501-970 Porto Alegre, RS, Brazil.

Authors’ Contribution

---

Pedro Ivo C. C. Figueiredo: Conceptualization, Data curation, Formal analysis, Methodology, Writing-original draft, Writing-review and editing.

Diogo T. Hashimoto: Methodology, Resources, Writing-review and editing.

Raquel B. Ariede: Data curation, Methodology.

André L. S. Zani: Formal analysis, Writing-review and editing.

Maikel Varal: Formal analysis, Writing-review and editing.

Luiz R. Malabarba: Conceptualization, Funding acquisition, Resources, Writing-review and editing.

Nelson J. R. Fagundes: Conceptualization, Formal analysis, Funding acquisition, Writing-original draft, Writing-review and editing.

Ethical Statement​

---

All fishes used in this study were collected in accordance with Brazilian laws, under a scientific collection license to PICCF (ICMBio-MMA 77839-1). The Ichthyology Laboratory at UFRGS has federal permission to keep specimens and tissues in a public collection under our care. Fish collections by lab members were approved by the University’s ethical committee (CEUA/UFRGS 24434).

Competing Interests

---

The author declares no competing interests.

How to cite this article

---

Figueiredo PICC, Hashimoto DT, Ariede RB, Zani ALS, Varal M, Malabarba LR, Fagundes NJR. Population genomics of Gymnogeophagus labiatus and G. lacustris reinforces the role of hydrographic isolation between evolutionary lineages (Cichlidae: Geophagini). Neotrop Ichthyol. 2025; 23(1):e240051. https://doi.org/10.1590/1982-0224-2024-0051

Copyright​

---

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Distributed under

Creative Commons CC-BY 4.0

© 2025 The Authors.

Diversity and Distributions Published by SBI

Accepted January 28, 2025 by Hernán López-Fernández

Submitted June 7, 2024

Epub April 18, 2025