Coding the Tree of Life: A New Era for Species Delimitation

Introduction: The Endless Struggle to Define Life’s Units

The observation of the natural world reveals a striking and pervasive phenomenon: life is not a continuous smear of variation but is organized into discrete clusters. When we walk through a forest, we see oak trees and maple trees, but we do not see a continuous gradation of forms linking them. When we observe the birds at a feeder, we distinguish the cardinal from the jay with ease. This discontinuity—the "lumpiness" of biological diversity—is the fundamental reality that allows the science of biology to exist. If life were a seamless continuum, we could not name organisms, we could not study their specific traits, and we could not conserve them. These discrete clusters are what we call species.

However, beneath this apparent order lies a chaotic and complex evolutionary process that has bedeviled scientists for centuries. The "Species Problem"—the difficulty in defining exactly what a species is and where the boundaries between them lie—is perhaps the oldest and most contentious debate in systematics. From the essentialist categories of Aristotle to the reproductive definitions of the mid-20th century, humanity has struggled to impose a rigid classification system on the fluid process of evolution.

Today, we stand at the threshold of a new era in this enduring quest. The advent of genomic science has provided us with a lens of unprecedented power. We are no longer limited to observing the external phenotype—the shape of a beak, the number of petals, or the pattern of spots. We can now peer directly into the molecular history of lineages, reading the distinct signatures of evolutionary divergence written in the code of life itself. This revolution is known as Genomic Species Delimitation.

This report provides an exhaustive examination of this transforming field, anchored by the landmark review "A Genomic Perspective on Species Delimitation" by Sonal Singhal and colleagues (2025), published in the Annual Review of Ecology, Evolution, and Systematics.¹ We will traverse the historical and theoretical landscapes that led to this moment, explore the sophisticated statistical models like the Multispecies Coalescent that underpin modern analysis, and immerse ourselves in detailed case studies ranging from the wind-swept surface of the open ocean to the microscopic battles within soil bacteria. We will find that while genomics offers powerful new tools for discovery, it also resurrects old philosophical demons regarding the nature of biological reality, forcing us to ask: Are we discovering species, or are we creating them?

The Historical Context: From Essence to Process

To fully appreciate the genomic revolution, one must situate it within the arc of taxonomic history. For centuries, species were defined by Typology. A species was an "essence," a fixed ideal. Variation was seen as noise or imperfection. If a specimen matched the "type," it belonged to the species. This Morphological Species Concept served humanity well for cataloging the macroscopic world but crumbled when faced with cryptic variation and sexual dimorphism.

The 20th century brought the Modern Synthesis, uniting Darwinian evolution with Mendelian genetics. Out of this emerged the Biological Species Concept (BSC), championed by Ernst Mayr. The BSC shifted the focus from pattern (what they look like) to process (do they mate?). A species became defined as a group of actually or potentially interbreeding natural populations which are reproductively isolated from other such groups.³ This was a profound conceptual leap, emphasizing that the "glue" holding a species together was gene flow—the exchange of genetic material.

Yet, the BSC proved difficult to operationalize. How does one test reproductive isolation between two populations of frogs living on different continents? How does one apply it to asexual bacteria or to plants that hybridize freely yet remain distinct?

The Genomic Era has not discarded these concepts but has subsumed them into a more data-rich framework. Genomic species delimitation does not rely on a single criterion. Instead, it views speciation as a continuum—a "Grey Zone" where lineages gradually accumulate differences.² By analyzing genome-wide data, we can detect the cessation of gene flow, the accumulation of divergent alleles, and the demographic history of splitting populations long before these changes manifest in external morphology. The genomic perspective allows us to quantify the process of speciation rather than just describing its end products.

The Theoretical Engine: The Multispecies Coalescent

The transition from traditional morphological taxonomy to genomic delimitation is not merely a change in data; it is a change in the underlying mathematical model of reality. At the heart of this new paradigm lies the Multispecies Coalescent (MSC) model. To understand modern species delimitation, one must understand the MSC.⁶

The Discordance of Gene Trees and Species Trees

In the early days of molecular phylogenetics, researchers often sequenced a single gene—typically a mitochondrial gene like COI—and assumed that the evolutionary tree of that gene (the gene tree) was identical to the evolutionary tree of the species (the species tree). If the gene tree showed that Species A and Species B were sisters, it was concluded that the species were sisters.

Genomics revealed this to be a dangerous simplification. As we began to sequence hundreds or thousands of loci across the genome, a startling pattern emerged: different genes often told different histories. Gene A might suggest Humans and Chimps are closest relatives; Gene B might suggest Chimps and Gorillas are closest. This phenomenon is known as gene tree discordance.⁷

This discordance is not error; it is a biological reality caused by a phenomenon called Incomplete Lineage Sorting (ILS).

Visualizing Incomplete Lineage Sorting (ILS)

Imagine the evolutionary process as a giant Pachinko machine or a game of chance involving colored marbles.¹⁰

• The Ancestor: Picture an ancestral population as a container holding thousands of marbles. These marbles represent the different alleles (variants) of a specific gene present in that population. Let’s say there are Red marbles and Blue marbles.

• The Split: When this ancestral species splits into two daughter lineages (Speciation Event 1), the marbles tumble down into two new containers. This process is random (genetic drift). One daughter species might get mostly Red marbles, the other mostly Blue.

• The Sorting: If the daughter lineages persist for a long time before splitting again, the marbles "sort." Genetic drift will eventually cause one lineage to become fixed for Red and the other for Blue.

• The Incomplete Sort: However, if the time between speciation events is short, or if the ancestral population size was huge (meaning lots of marbles to sort), the lineages might not finish sorting before the next split (Speciation Event 2) occurs.

Consider three species: A, B, and C. They evolved from a common ancestor. First, the ancestor split into the A/B lineage and the C lineage. Later, the A/B lineage split into A and B.

If the ancestral A/B population still contained both Red and Blue marbles when it split, Species A might randomly inherit the Red allele, and Species B might inherit the Blue allele. Crucially, Species C (which split off earlier) might also have inherited the Blue allele.

A researcher sequencing this gene would see that Species B and Species C share the Blue allele, while A has the Red one. They would draw a gene tree grouping B and C together. But the true species history groups A and B together. The gene tree clashes with the species tree because the ancestral lineage failed to "sort" its alleles completely before the second speciation event.7

The MSC as the Solution

The Multispecies Coalescent model was developed to make sense of this chaos. It acknowledges that gene trees are embedded within species trees, constrained by the species' history but free to vary within those boundaries.¹²

Instead of trying to force all genes to agree, the MSC uses the variation among gene trees as information. The frequency of discordant gene trees tells us about the parameters of the species tree.

• Branch Lengths: Short branches (rapid speciation) produce more ILS and thus more discordance.

• Population Sizes: Large ancestral populations maintain more variation (more marbles), leading to more ILS.

By integrating over thousands of gene trees, the MSC can statistically infer the true species tree topology and the divergence times, effectively filtering out the noise of ILS to reveal the underlying evolutionary signal. This model is the foundation for validation methods in genomic species delimitation, allowing researchers to ask: "Is the genetic variation we see consistent with two species diverging, or is it just the noisy variation of a single large population?".⁹

Methodological Frameworks: Discovery and Validation

Singhal et al. (2025) outline a rigorous, two-step workflow that has become the standard for genomic species delimitation: Species Discovery followed by Species Validation.² This separation is crucial for maintaining scientific objectivity and avoiding circular reasoning.

Step 1: Species Discovery (Hypothesis Generation)

The discovery phase involves exploring the data without rigid a priori assumptions to identify "primary species hypotheses" or Operational Taxonomic Units (OTUs). The goal is to find clusters of individuals that are genetically distinct enough to warrant further testing.

1. Clustering Algorithms (Structure / Admixture):

These methods, rooted in population genetics, do not use phylogenetic trees. Instead, they view individuals as mixtures of genetic contributions from K unknown ancestral populations.

• The Logic: The algorithm attempts to minimize deviations from Hardy-Weinberg equilibrium and Linkage Equilibrium within clusters. If a group of individuals breeds randomly (a species), their alleles should be shuffled in predictable ways. If there are barriers to breeding (two species), distinct associations of alleles will persist.

• The Output: The researcher runs the model for different values of K (e.g., K=1 to K=10). The software calculates the likelihood of the data for each K. If the likelihood peaks at K=3, it suggests there are three distinct genetic clusters.¹³

2. Distance-Based Methods (ASAP):

Methods like Assemble Species by Automatic Partitioning (ASAP) analyze the pairwise genetic distances between all sequences.

• The Barcode Gap: They search for a "gap" in the distribution of distances—a threshold where intra-specific variation (differences between individuals of the same species) ends and inter-specific divergence (differences between species) begins.

• Advantage: These methods are computationally fast and do not require complex modeling, making them excellent for initial screening of large datasets.¹⁵

Step 2: Species Validation (Hypothesis Testing)

Once putative species are identified, they must be subjected to rigorous statistical testing. This is where the Multispecies Coalescent shines. The question shifts from "What clusters exist?" to "Do these clusters behave like independent evolutionary lineages?"

1. BPP (Bayesian Phylogenetics and Phylogeography):

BPP is the titan of validation methods. Developed by Yang and Rannala, it uses a full Bayesian implementation of the MSC.2

• The Mechanism: BPP uses a Reversible-Jump Markov Chain Monte Carlo (rjMCMC) algorithm.

• Imagine a "state space" that contains all possible species delimitation models. One state might be "A, B, and C are separate species." Another state might be "A and B are one species; C is separate."

• The MCMC robot wanders through this space, visiting different models. It spends more time in models that have a higher probability of producing the observed genomic data.

• The "reversible jump" capability allows the robot to jump between dimensions—moving from a model with 2 species (fewer parameters) to a model with 3 species (more parameters).¹⁸

• The Result: After millions of steps, the researcher checks where the robot spent its time. If it spent 99% of the simulation in the "3 species" model, the posterior probability of that delimitation is 0.99.

2. Bayes Factor Delimitation (BFD):*

This method compares specific delimitation models using Bayes Factors, which represent the strength of evidence for one hypothesis over another.

• The Process: Researchers explicitly define competing models (e.g., Model 1: "West and East populations are one species"; Model 2: "West and East are distinct"). They use the SNAPP algorithm to calculate the marginal likelihood of each model.

• The Comparison: The difference in marginal likelihoods constitutes the Bayes Factor. A sufficiently high BF provides decisive support for one model, allowing researchers to mathematically justify splitting or lumping taxa.¹⁶

3. The Genealogical Divergence Index (GDI):

Recognizing that BPP can be "too sensitive" (detecting population structure rather than species), the GDI was developed as a heuristic filter.

• The Index: It combines estimates of population size (theta) and divergence time (tau) into a single index ranging from 0 to 1.

• Interpretation:

• GDI < 0.2: Likely a single species (panmixia).

• 0.2 < GDI < 0.7: The "Grey Zone" (ambiguous structure).

• GDI > 0.7: Strong support for distinct species.

• Role: GDI is often used to "validate the validation," ensuring that the species detected by BPP are biologically substantial and not just transient population ripples.²¹

Case Study I: The Open Ocean – Siphonophores and the Myth of the Cosmopolitan

To illustrate the transformative power of these methods, Singhal et al. (2025) point to the open ocean, a habitat often assumed to be a single, barrier-free environment. Here, the re-evaluation of siphonophores (Cnidaria: Hydrozoa) serves as a dramatic example of how genomics can overturn centuries of established taxonomy.

The Organism: A Colonial Existence

Siphonophores are among the most complex and alien life forms on Earth. They are not single individuals but colonies of specialized zooids—clones derived from a single fertilized egg that function as organs. In a Portuguese Man o' War (Physalia), one zooid becomes the gas-filled float (pneumatophore), others become tentacles (dactylozooids), others digest food (gastrozooids), and others handle reproduction (gonozooids). They drift passively, driven by wind and currents.²³

The Historical "Lumping"

For much of the 20th century, the taxonomy of Physalia was dominated by the work of A.K. Totton (1960). Totton reasoned that because the open ocean lacks obvious physical barriers (like mountains or rivers), and because Physalia drift thousands of kilometers, there could be no geographic isolation. Consequently, he synonymized dozens of previously described species into a single, global species: Physalia physalis.²⁵ This "cosmopolitan" hypothesis prevailed for decades.

The Genomic Revelation: Church et al. (2025)

Singhal et al. highlight the study by Church et al. (2025), "Population genomics of a sailing siphonophore reveals genetic structure in the open ocean," as a paradigmatic shift.27

Church and colleagues assembled a global collection of Physalia and applied whole-genome sequencing. The results were unequivocal: the "single species" hypothesis was wrong.

1. Discovery of Deep Lineages:

Using genomic clustering, they identified multiple distinct lineages that had been evolving independently for millions of years. They resurrected three species and described a new one:

• Physalia physalis: Restricted primarily to the North Atlantic.

• Physalia utriculus: Found in the Indo-Pacific.

• Physalia megalista: Another distinct lineage.

• Physalia minuta: A newly described, smaller species.²⁴

2. The Mechanism: Sailing Dynamics as Barriers

How does speciation occur in a featureless ocean? The validation step of the study integrated genomic data with oceanographic modeling.

The pneumatophore of Physalia is an aerodynamic sail. Crucially, the colonies are dimorphic in their sailing handedness: "left-handed" colonies drift 45 degrees to the left of the wind, and "right-handed" colonies drift to the right.

Church et al. hindcast the drift trajectories of thousands of specimens. They found that the interaction of global wind patterns (Trade Winds, Westerlies) and ocean currents (Gulf Stream, Gyres) creates invisible but formidable barriers. A Physalia in the North Atlantic Gyre is effectively trapped there by the physics of its drift. It will never meet or mate with a Physalia in the South Atlantic or Pacific.24

The genomic boundaries perfectly matched these "drift divides." The study demonstrated that physics acts as a reproductive barrier in the open ocean, driving allopatric speciation just as effectively as a mountain range on land.

The Nanomia Complex: Genome Size Evolution

The review also discusses Nanomia, another siphonophore genus. Traditionally containing only two species (N. bijuga and N. cara), genomic analysis revealed cryptic diversity.

• Genome Size Variation: Flow cytometry integrated with phylogenomics showed a shocking disparity. Nanomia septata and N. cara have massive genomes (approx. 1.5 - 1.7 GB). In contrast, N. bijuga and a newly discovered cryptic species (Nanomia sp. 1) have undergone a secondary reduction in genome size to approx. 0.7 GB.²⁸

• Implication: This suggests that speciation in this group is accompanied by drastic reorganization of genomic architecture. The "cryptic" species were not just genetically divergent; they were fundamentally different in their cellular biology.

These siphonophore studies exemplify how genomic delimitation corrects taxonomic undersplitting (lumping). By ignoring the subtle signals of divergence and relying on coarse morphology, traditional taxonomy masked the true diversity and evolutionary history of the pelagic realm.

Case Study II: The Terrestrial Realm – Birds, Lizards, and the Complexity of Gene Flow

While the ocean revealed hidden species, the terrestrial realm provides examples of the opposite problem: how genomics complicates the simple picture of "distinct" species, particularly in groups like birds and lizards where hybridization is common.

The Redpoll Finch Paradox: Diversity Without Divergence

The Redpoll finches (Acanthis spp.) are a classic taxonomic headache. Birders distinguish between Common Redpolls, Hoary Redpolls, and Lesser Redpolls based on plumage, bill size, and body mass. Under the Morphological Species Concept, they are distinct.

However, Singhal et al. (2025) discuss genomic studies showing that these "species" are genomically homogenized. Sequencing reveals panmixia (random mating) across almost the entire genome. There is no deep phylogenetic split separating a Hoary Redpoll from a Common Redpoll.¹

The Mechanism: Islands of Divergence

How can they look so different if they are genetically the same? The answer lies in selection on specific genomic regions. The phenotypic differences are likely controlled by a few "supergenes" or islands of divergence—small clusters of genes involved in beak formation and pigmentation.

• Natural selection maintains these specific alleles in different environments (e.g., larger beaks for harder seeds in one range, smaller beaks in another).

• However, the rest of the genome flows freely between populations through hybridization.

This presents a philosophical challenge for delimitation.

• If a species is defined by genome-wide isolation, Redpolls are one species.

• If a species is defined by adaptive maintenance of phenotype, they might be multiple.The review argues that genomic delimitation in such cases does not give a binary "yes/no" answer but rather elucidates the mechanism of divergence (selection with gene flow).30

Songbirds and the "Ghost of Introgression"

Introgression (the movement of genes from one species to another via hybridization) is rampant in birds. This poses a severe risk for species delimitation methods that assume strictly bifurcating trees.

• Mito-Nuclear Discordance: Birds often capture the mitochondrial genome of a sister species. A study of "Leopard Frogs" (comparable to avian cases) showed that mitochondrial DNA often delimits "species" that genomic nuclear data reveals are just admixed populations.³¹

• Gene Flow as a Proxy: Singhal et al. emphasize that in these systems, estimating gene flow is a critical validation step.

• Validation tools must incorporate migration parameters (e.g., using the Isolation with Migration model).

• If the model shows high, continuous gene flow, the "separate species" hypothesis is weakened.

• If gene flow is zero or historical only, the species status is robust.

• Insight: Genomic delimitation uses gene flow estimates as a proxy for reproductive isolation. Since we cannot force-mate every bird population in the lab, we use the genome to ask, "Have they been mating in the wild?".²

Lizards: Niche vs. Genome

In lizards, Singhal’s own work (e.g., on Carlia and Xantusia) illustrates how genomic data can be integrated with environmental modeling.

• Isolation by Distance (IBD): In lizards with low dispersal ability, populations often show immense genetic structure simply because they don't move far. A genomic model might identify every mountain top population as a "species."

• Integrative Solution: By testing for IBD and comparing genomic divergence to ecological divergence, researchers can avoid oversplitting. If the genetic difference is purely a function of distance (linear relationship) rather than a sharp break at a barrier, it is likely just population structure.³²

Case Study III: The Microscopic Realm – Bacteria and the Definition of "Species"

In the microbial world, the concept of a species reaches its breaking point. Bacteria are asexual; they reproduce clonally. They also engage in Horizontal Gene Transfer (HGT), swapping genes across vast phylogenetic distances. The Biological Species Concept (reproductive isolation) is theoretically inapplicable. Yet, bacteriologists need names to communicate about pathogens and ecological agents.

The Streptomyces Challenge

Singhal et al. (2025) and associated literature (e.g., Wang et al. 2022) use the genus Streptomyces—soil bacteria famous for antibiotic production—to illustrate the genomic redefinition of the microbial species.³³

Traditional Bacteriology:

Historically, bacterial species were defined by 16S rRNA similarity. If two strains shared >97% sequence identity at this one gene, they were the same species. This was a crude yardstick, lumping together organisms with vastly different ecologies.

Genomic Delimitation (ANI):

The current standard is Average Nucleotide Identity (ANI) calculated from whole genomes. The threshold is typically set at 95-96%. If two genomes are >95% identical, they are the same species.35

Ecological Speciation in the Micro-Scale

However, the Singhal review pushes beyond arbitrary thresholds to a process-based view. Wang et al. (2022) investigated Streptomyces olivaceus and found it contained two distinct lineages driven by Habitat Adaptation.³⁴

1. Free-Living (FL) Lineage: Adapted to the soil environment. These genomes were large, containing extensive gene clusters for iron scavenging (siderophores) needed in the competitive, nutrient-poor soil.

2. Insect-Associated (IA) Lineage: Adapted to live symbiotically within insects. These genomes were streamlined (smaller) but possessed specific genes for utilizing insect-derived substrates like sialic acid and glycogen.

The Pan-Genome Driver:

The speciation process here was driven by the accessory genome (the "pan-genome"). The "Core Genome" (genes shared by all Streptomyces) remained relatively similar. But the gain/loss of specific functional gene clusters (the accessory genes) allowed the IA lineage to colonize a new niche (the insect gut).

Once inside the insect, the IA lineage became spatially isolated from the FL lineage in the soil. This ecological isolation reduced gene flow, leading to speciation.

Insight: In bacteria, genomic species delimitation is not just about counting SNPs; it is about analyzing the functional content of the genome to identify the ecological shifts that drive lineage divergence. The "species" is an ecological unit, defined by its niche and the genomic toolkit required to occupy it.³⁷

The Great Debate: The Peril of Oversplitting

The power of genomic data to detect fine-scale structure has led to a crisis in taxonomy known as Oversplitting. This is a central theme in the Singhal et al. review and a subject of intense debate in the field.⁹

The Conflict: Structure vs. Speciation

The Multispecies Coalescent (MSC) is designed to detect genetic structure—deviations from panmixia.

• The Problem: All species have population structure. Humans have population structure. If you feed the genomes of a villager from the Andes and a villager from the Himalayas into BPP, the program will correctly detect that they are genetically distinct lineages with no recent gene flow. It might assign them a high posterior probability of being "distinct."

• The Interpretation: Does that make them separate species? Biologically, no. They are potentially interbreeding populations of the same species.

• The Trap: Because the MSC interprets any cessation of gene flow as a "speciation" branch, it tends to elevate populations to species status. As we sequence more genomes, our statistical power to detect these minor splits increases. We risk a future where every isolated population is named a species, inflating biodiversity counts by orders of magnitude and rendering the term "species" meaningless.³⁹

The Critique: Sukumaran & Knowles (2017)

The review cites the influential critique by Sukumaran and Knowles, who argued that the MSC cannot distinguishing between structure (population dynamics) and speciation (evolutionary independence). They warned that delimitation results from programs like BPP should be treated as hypotheses of lineages, not definitive species, unless other data supports them.⁹

The Solution: Integrative Taxonomy

To combat oversplitting, Singhal et al. (2025) advocate for Integrative Taxonomy. This approach treats genomic data as the backbone of the analysis but requires corroboration from other "axes of divergence".⁴¹

The Integrative Workflow:

1. Genomic Discovery: Use structure/MSC to identify potential lineages (OTUs).

2. Phenotypic Check: Do these lineages look different? (Morphology).

3. Ecological Check: Do they occupy different niches? (Ecology).

4. Geographic Context:

5. Sympatry: If the lineages live in the same place (sympatry) and remain distinct, they are Good Species (strong reproductive isolation).

6. Allopatry: If the lineages are separated by a mountain range (allopatry) and look identical, they are likely Populations (structure due to isolation).

7. Parapatry: If they meet at a hybrid zone, the width of the zone relative to dispersal distance determines the status. Narrow hybrid zones suggest species; wide ones suggest populations.³²

Table 1: The Axes of Integrative Taxonomy

Data Source	Question Asked	Role in Delimitation	Risk of Error
Genomics (MSC)	Do lineages sort alleles independently?	Foundation/Backbone	Oversplitting (confusing populations for species)
Morphology	Do lineages look different?	Corroboration / Field ID	Undersplitting (missing cryptic species)
Ecology	Do lineages use different niches?	Explains mechanism	Plasticity (phenotype changes with environment, not genes)
Geography	Are lineages physically isolated?	Context for Gene Flow	Isolation by Distance (confusing distance for divergence)

Conclusion: The Future of Taxonomy

The review by Singhal et al. (2025) delineates a field in transition. We have moved from the "Pattern Era" of morphology, through the "Barcode Era" of single genes, and have arrived at the Process Era of genomics.

We now understand that a species is not a static type. It is a dynamic evolutionary lineage, a thread in the tapestry of life that maintains its integrity through time against the forces of erosion (hybridization) and fragmentation (drift).

• Discovery is Easier: We can now find species that were invisible to us, like the cryptic Physalia lineages hiding in plain sight on the ocean waves, driven apart by the physics of wind.

• Definition is Harder: We are forced to confront the blurry reality of biology. The Redpoll finches teach us that a "species" can be a genomic ghost, maintained by a sliver of adaptive DNA. The Streptomyces teach us that a species can be defined by the tools it carries to survive in an insect gut.

• Responsibility is Higher: The power to detect difference comes with the responsibility to interpret it wisely. The risk of oversplitting threatens to break the utility of taxonomy. The solution lies not in better algorithms, but in better biology—integrating the genomic signal with the ecological and phenotypic reality of the organism.

For the undergraduate researcher, the lesson is this: Data is not the answer; it is the question. The genome provides the most detailed historical record of life ever available, but it requires a biologist to read it. As we face a global biodiversity crisis, with species going extinct before they are even named, the speed and accuracy of genomic species delimitation have never been more critical. We are building the library of life, and genomics is the pen with which we write the names.

Key Concepts Glossary

• Allopatric Speciation: Speciation that occurs when biological populations of the same species become geographically isolated from each other.

• Coalescent Theory: A retrospective model of population genetics that traces the ancestry of gene copies back in time to a single common ancestor.

• Cryptic Species: Two or more distinct species that were classified as a single species because they are morphologically indistinguishable.

• Gene Flow: The transfer of genetic material from one population to another.

• Genomic Island of Divergence: A small region of the genome that is highly differentiated between species, often containing genes under strong selection, while the rest of the genome remains similar due to gene flow.

• Haplotype: A group of genes within an organism that was inherited together from a single parent.

• Holotype: A single physical example (or illustration) of an organism, known to have been used when the species (or lower-ranked taxon) was formally described.

• Introgressive Hybridization (Introgression): The movement of a gene (gene flow) from one species into the gene pool of another by the repeated backcrossing of an interspecific hybrid with one of its parent species.

• Mito-nuclear Discordance: A situation where the evolutionary history told by mitochondrial DNA differs from that told by nuclear DNA.

• Monophyletic: A group of organisms that consists of all the descendants of a common ancestor.

• Parapatric Speciation: Speciation where the zones of two diverging populations are only partially separated or adjacent.

• Polyphyletic: A group of organisms derived from more than one common evolutionary ancestor or ancestral group and therefore not suitable for placing in the same taxon.

• Sympatric Speciation: The evolution of a new species from a surviving ancestral species while both continue to inhabit the same geographic region.

• Synonymy: The state of being a synonym; in taxonomy, when multiple names have been applied to the same species, the later names are synonymized with the first valid name.

• Taxon (pl. Taxa): A group of one or more populations of an organism or organisms seen by taxonomists to form a unit (e.g., species, genus, family).

Works cited

1. A Genomic Perspective on Species Delimitation | Request PDF - ResearchGate, accessed January 12, 2026, https://www.researchgate.net/publication/395252778_A_Genomic_Perspective_on_Species_Delimitation

2. A Genomic Perspective on Species Delimitation - Annual Reviews, accessed January 12, 2026, https://www.annualreviews.org/doi/pdf/10.1146/annurev-ecolsys-102723-055311

3. (PDF) Species delimitation - ResearchGate, accessed January 12, 2026, https://www.researchgate.net/publication/343920814_Species_delimitation

4. Species Delimitation - SMILE, accessed January 12, 2026, https://www.smile.cnrs.fr/documents/chapter_5.5_YangRannala.pdf

5. A Genomic Perspective on Species Delimitation - Annual Reviews, accessed January 12, 2026, https://www.annualreviews.org/content/journals/10.1146/annurev-ecolsys-102723-055311

6. Hierarchical Heuristic Species Delimitation Under the Multispecies Coalescent Model with Migration | Systematic Biology | Oxford Academic, accessed January 12, 2026, https://academic.oup.com/sysbio/article/73/6/1015/7740481

7. Multispecies coalescent process - Wikipedia, accessed January 12, 2026, https://en.wikipedia.org/wiki/Multispecies_coalescent_process

8. The multi-species coalescent model and species tree inference - University of Washington, accessed January 12, 2026, https://faculty.washington.edu/leache/wordpress/wp-content/uploads/2020/03/2020RannalaSpeciestree.pdf

9. Multispecies coalescent delimits structure, not species - PNAS, accessed January 12, 2026, https://www.pnas.org/doi/10.1073/pnas.1607921114

10. Incomplete lineage sorting through Pachinko – a visual analogy - The G-cat, accessed January 12, 2026, https://theg-cat.com/2021/03/01/incomplete-lineage-sorting-through-pachinko-a-visual-analogy/

11. The Multilocus Multispecies Coalescent: A Flexible New Model of Gene Family Evolution, accessed January 12, 2026, https://www.biorxiv.org/content/10.1101/2020.05.07.081836v4.full-text

12. Multispecies coalescent and its applications to infer species phylogenies and cross-species gene flow - PMC - NIH, accessed January 12, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC8692950/

13. Species Delimitation Using Genomic Data: Options and Limitations - PMC - NIH, accessed January 12, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC11974488/

14. Evaluating Species Delimitation Methods in Radiations: The Land Snail Albinaria cretensis Complex on Crete | Systematic Biology | Oxford Academic, accessed January 12, 2026, https://academic.oup.com/sysbio/article-abstract/71/2/439/6310756

15. Unexpected diversity and novel lineages in the cosmopolitan genus Nanomia (Cnidaria: Siphonophorae: Physonectae) - Frontiers, accessed January 12, 2026, https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2024.1421514/full

16. Species boundaries in the messy middle—A genome‐scale validation of species delimitation in a recently diverged lineage of coastal fog desert lichen fungi - PubMed Central, accessed January 12, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC8717302/

17. Species Delimitation with Gene Flow | Systematic Biology - Oxford Academic, accessed January 12, 2026, https://academic.oup.com/sysbio/article/66/5/799/2726792

18. The Spectre of Too Many Species - PMC - NIH, accessed January 12, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC6292489/

19. Bayesian species delimitation using multilocus sequence data - PNAS, accessed January 12, 2026, https://www.pnas.org/doi/10.1073/pnas.0913022107

20. Species boundaries to the limit: validating species delimitation methods is critical to avoid taxonomic inflation in the genomic era | bioRxiv, accessed January 12, 2026, https://www.biorxiv.org/content/10.1101/2022.12.28.522089v1.full-text

21. Establishing species boundaries in Bornean geckos | Biology Letters | The Royal Society, accessed January 12, 2026, https://royalsocietypublishing.org/rsbl/article/20/8/20240157/63641/Establishing-species-boundaries-in-Bornean

22. Spectre of Too Many Species | Systematic Biology - Oxford Academic, accessed January 12, 2026, https://academic.oup.com/sysbio/article/68/1/168/5049398

23. Apolemia uvaria collected from the surface water. Top left: A whole... - ResearchGate, accessed January 12, 2026, https://www.researchgate.net/figure/Apolemia-uvaria-collected-from-the-surface-water-Top-left-A-whole-colony-with-the-stem_fig2_279890104

24. Physalia mikazuki sp. nov. (Phylum Cnidaria; class Hydrozoa) blown into Japan's northeast (Tohoku) at the whim of marine ecosystem change - Frontiers, accessed January 12, 2026, https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2025.1653958/full

25. Global genomics of man-o'-war (Physalia) reveals ocean surface biodiversity, accessed January 12, 2026, https://qmro.qmul.ac.uk/xmlui/bitstream/handle/123456789/108315/Mendoza%20Population%20genomics%20of%202025%20Published.pdf.pdf?sequence=5

26. From surfacing to stranding: The origins and dispersal dynamics of a neustonic siphonophore - bioRxiv, accessed January 12, 2026, https://www.biorxiv.org/content/biorxiv/early/2025/08/15/2025.08.13.669909.full.pdf

27. Population genomics of a sailing siphonophore reveals genetic structure in the open ocean | Request PDF - ResearchGate, accessed January 12, 2026, https://www.researchgate.net/publication/392849093_Population_genomics_of_a_sailing_siphonophore_reveals_genetic_structure_in_the_open_ocean

28. Map of Nanomia observations listed in Supplementary Table S1 - ResearchGate, accessed January 12, 2026, https://www.researchgate.net/figure/Map-of-Nanomia-observations-listed-in-Supplementary-Table-S1_fig2_384964925

29. (PDF) A Chromosome-Scale Genome of Nanomia septata Reveals Extensive Rearrangement But No Clear Driver of the Unique Colony-Level Organization of Siphonophores - ResearchGate, accessed January 12, 2026, https://www.researchgate.net/publication/391533736_A_Chromosome-Scale_Genome_of_Nanomia_septata_Reveals_Extensive_Rearrangement_But_No_Clear_Driver_of_the_Unique_Colony-Level_Organization_of_Siphonophores

30. A Genomic Perspective on Species Delimitation - R Discovery - Researcher.Life, accessed January 12, 2026, https://discovery.researcher.life/article/a-genomic-perspective-on-species-delimitation/fb8db7fba3af3bac9e695c91d47ed225

31. Distinguishing species boundaries from geographic variation - PNAS, accessed January 12, 2026, https://www.pnas.org/doi/10.1073/pnas.2423688122

32. (PDF) Integrative taxonomy and geographic sampling underlie successful species delimitation - ResearchGate, accessed January 12, 2026, https://www.researchgate.net/publication/350827704_Integrative_taxonomy_and_geographic_sampling_underlie_successful_species_delimitation

33. Habitat Adaptation Drives Speciation of a Streptomyces Species with Distinct Habitats and Disparate Geographic Origins | mBio - ASM Journals, accessed January 12, 2026, https://journals.asm.org/doi/10.1128/mbio.02781-21

34. Habitat Adaptation Drives Speciation of a Streptomyces Species with Distinct Habitats and Disparate Geographic Origins - NIH, accessed January 12, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC8749437/

35. Exclusivity offers a sound yet practical species criterion for bacteria despite abundant gene flow - PubMed Central, accessed January 12, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC6171291/

36. Challenges in the Identification of Environmental Bacterial Isolates from a Pharmaceutical Industry Facility by 16S rRNA Gene Sequences - MDPI, accessed January 12, 2026, https://www.mdpi.com/2673-8856/5/3/33

37. Interaction between Sugarcane and Environmental Stressors - MDPI, accessed January 12, 2026, https://mdpi-res.com/bookfiles/book/10042/Interaction_between_Sugarcane_and_Environmental_Stressors_From_Identification_to_Molecular_Mechanism.pdf?v=1735524501

38. Habitat adaptation drives speciation of a Streptomyces species with distinct habitats and disparate geographic origins - Kérwá, accessed January 12, 2026, https://www.kerwa.ucr.ac.cr/items/581b35c2-de6d-412e-a178-ad060f248677

39. Incorporating the speciation process into species delimitation - PMC - PubMed Central - NIH, accessed January 12, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC8118268/

40. Incorporating the speciation process into species delimitation | PLOS Computational Biology, accessed January 12, 2026, https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1008924

41. Phylogenomic Species Delimitation Dramatically Reduces Species Diversity in an Antarctic Adaptive Radiation - the NOAA Institutional Repository, accessed January 12, 2026, https://repository.library.noaa.gov/view/noaa/54301/noaa_54301_DS1.pdf

42. Integrative taxonomy | Naturalis, accessed January 12, 2026, https://www.naturalis.nl/en/integrative-taxonomy