Molecular Phylogeny of Monitor Lizards (Ast, 2001)

📖 21 min read~3731 words

This is an original editorial summary prepared by the Komodo Guide team; it is not a reproduction of the source article. Readers seeking the primary data are encouraged to consult the paper directly. In 2001, Jennifer C. Ast of the University of Michigan published a molecular-phylogenetic study that used mitochondrial DNA sequences to re-examine evolutionary relationships across the superfamily Varanoidea. The results reshuffled accepted ideas about monitor lizard subgenera, confirmed that Varanus komodoensis belongs to an Indo-Australian radiation, and raised new questions about how molecular evidence squares with the fossil record — questions that subsequent palaeontology has continued to refine.

Table of Contents

• Quick Facts
• Paper Overview
• The Molecular Toolkit
• Three Clades of Varanus
• Where the Komodo Dragon Fits
• Subgenera Under Scrutiny
• Molecular Versus Fossil Evidence
• Myths vs Facts
• Key Takeaways
• Frequently Asked Questions
• Sources & Further Reading

Quick Facts

Detail	Information
Full citation	Ast, J.C. (2001). "Mitochondrial DNA evidence and evolution in Varanoidea (Squamata)." Cladistics 17: 211–226.
Author affiliation	Museum of Zoology, University of Michigan, Ann Arbor
DOI	10.1111/j.1096-0031.2001.tb00118.x
Primary gene markers	ND2 (NADH dehydrogenase subunit 2) plus flanking tRNA genes; partial 16S rRNA and COI
Alignment length	More than 2,000 nucleotide positions
Analytical method	Maximum parsimony (single most parsimonious tree recovered)
Outgroup taxa	Heloderma spp. and Lanthanotus borneensis
Major result	Three-clade structure within Varanus: African, Indo-Asian, and Indo-Australian
Komodo dragon sister taxon	Varanus varius (lace monitor, eastern Australia)
Subgeneric impact	Subgenus Varanus s.s. found polyphyletic; Odatria found paraphyletic

Paper Overview

By the close of the twentieth century, the evolutionary relationships of monitor lizards had been reconstructed primarily through anatomical comparisons — bone shapes, scale counts, haemipenis morphology, and similar structural characters. These morphological analyses produced hypotheses about which species were most closely related, but they also generated persistent disagreements, particularly about the validity of the subgenera into which the roughly sixty recognised Varanus species had been parcelled. Ast's 2001 study introduced a wholly independent line of evidence: sequences of mitochondrial DNA, drawn from multiple genes and analysed with cladistic parsimony methods. By doing so, it offered the most explicitly molecular phylogenetic reconstruction of varanoid lizards available at that time and became a foundational reference for subsequent work on monitor evolution.

The scope of the study extended beyond Varanus alone. Varanoidea as a superfamily includes not just the monitors but also the Gila monster and related helodermatids (Heloderma spp.) and the enigmatic earless monitor lizard of Borneo, Lanthanotus borneensis. Ast used these two lineages as outgroups — evolutionary reference points outside the monitor genus itself — which allowed her to root the Varanus tree and determine the direction of evolutionary change. The monophyly of Varanoidea as a whole — meaning that helodermatids, Lanthanotus, and Varanus are all descended from a single common ancestor — was confirmed by the analysis, as was the sister-group relationship between Varanus and Lanthanotus. These broad conclusions supported prior morphological work; the sharper disagreements lay within Varanus itself.

The Molecular Toolkit

Mitochondrial DNA became a workhorse of evolutionary biology in the 1990s for several reasons. First, it is inherited maternally and does not undergo the recombination that shuffles nuclear chromosomes, so it accumulates mutations in a relatively clock-like fashion. Second, its genes tend to evolve faster than many nuclear genes, generating enough variation to distinguish species and genera within the timeframes relevant to lizard diversification. Third, by 2001 the laboratory techniques for amplifying and sequencing specific mitochondrial regions — PCR followed by Sanger sequencing — were sufficiently routine to permit broad taxonomic surveys.

Ast focused primarily on the gene encoding NADH dehydrogenase subunit 2 (ND2), together with a suite of flanking transfer-RNA genes. ND2 sits in a region of the mitochondrial genome that evolves at a useful intermediate rate: fast enough to distinguish recently diverged species, slow enough that the signal is not completely overwritten in deeper parts of the tree. Additional sequence data from partial 16S ribosomal RNA and cytochrome oxidase subunit I (COI) supplemented the core dataset. Taken together, the alignment covered more than 2,000 nucleotide positions, a substantial dataset for its era.

The analytical method was maximum parsimony, the dominant approach in systematic biology through the 1990s. Parsimony algorithms seek the evolutionary tree that requires the fewest total changes across all characters — in this case, nucleotide substitutions. When Ast ran her parsimony analysis across the combined mitochondrial dataset, a single most parsimonious tree emerged, meaning that one topology fit the data better than all alternatives by the criterion she applied. This clean result gave the findings unusual authority for a molecular study of its generation.

Three Clades of Varanus

The most architecturally significant finding of Ast (2001) was the recovery of three major lineages within Varanus, which subsequent research has repeatedly corroborated. These three groups are geographically coherent, though not without internal complexity:

The African clade. African monitor species — including the Nile monitor (V. niloticus) and its relatives — emerged as the first branch off the Varanus stem, meaning they are the sister group to all other monitors. In phylogenetic language, the African species are the most basally divergent members of the genus. This placement accords with biogeographic hypotheses that situate the origin of Varanus in Africa or the Tethyan margins of the Old World, though the molecular data alone cannot settle the question of ancestral geography. The fossil-record evidence bearing on this question is addressed separately on the fossil record page.

The Indo-Asian clade. Species distributed across South and Southeast Asia — most prominently the water monitor V. salvator and relatives — form a second major group. Ast's analysis suggested this assemblage is paraphyletic with respect to some Indonesian taxa, meaning the Indo-Asian group does not form a perfectly closed evolutionary unit but instead bleeds into the Indo-Australian radiation geographically and phylogenetically. This paraphyly reflects the complex biogeographic history of the Sundaland region, where rising and falling sea levels during ice ages have repeatedly connected and isolated island archipelagos.

The Indo-Australian clade. A well-supported grouping of species centred on Australia and the Indo-Pacific islands constitutes the third lineage, and it is the one most directly relevant to understanding Varanus komodoensis. Within this clade, the large-bodied species of the gouldii group — the goannas of mainland Australia — and the smaller Australasian monitors placed in subgenus Odatria (pygmy monitors and their relatives) were resolved as distinct subgroups, though with a complex internal arrangement that challenged prior classification schemes.

Clade	Geographic Range	Phylogenetic Position	Representative Species
African	Sub-Saharan & North Africa	Sister to all other Varanus	V. niloticus, V. exanthematicus
Indo-Asian	South & Southeast Asia	Paraphyletic intermediate group	V. salvator, V. bengalensis
Indo-Australian	Australia, New Guinea, Pacific islands	Derived clade, includes Komodo dragon	V. komodoensis, V. varius, V. gouldii

Where the Komodo Dragon Fits

The placement of Varanus komodoensis within the Indo-Australian clade carries implications that extend well beyond a taxonomic rearrangement. Ast's parsimony tree resolves V. varius — the lace monitor of eastern Australia — as the closest living relative of the Komodo dragon. This sister-taxon relationship, inferred from mitochondrial sequence data, sits at the heart of debates about the origin and dispersal of the world's largest lizard.

If V. komodoensis is most closely related to an Australian species, then the ancestors of today's Komodo dragons almost certainly originated in or near Australia, rather than arriving in the Lesser Sundas from the Asian mainland. This molecular signal aligns strikingly with what the fossil record eventually revealed: Pliocene varanid remains from Queensland and other Australian localities have been interpreted as morphologically referable to V. komodoensis, suggesting the lineage was widespread across Australia before contracting westward. The discovery of giant varanid fossils from Timor and Flores — explored in depth in the Hocknull et al. (2009) review — corroborates a westward dispersal trajectory from mainland Australia through island chains to the current range in Komodo National Park.

A related implication concerns Varanus priscus (the extinct giant formerly known as Megalania). Subsequent researchers using Ast's phylogenetic framework have argued that large body size is a synapomorphy — a shared derived character — of the komodoensis–varius lineage rather than an isolated evolutionary experiment on Flores. In other words, both the living Komodo dragon and the extinct Australian giant may have inherited their imposing dimensions from a common large-bodied ancestor, rather than independently evolving gigantism in separate geographic locations. This hypothesis has profound implications for how we interpret insular dwarfism and giantism in island biogeography.

Subgenera Under Scrutiny

The traditional subgeneric classification of Varanus organised the approximately 80 monitor lizard species into geographic and morphological groupings, including Varanus (large, generalised species), Odatria (small Australasian pygmy monitors), and Euprepiosaurus (Indo-Pacific tree monitors). Ast's 2001 mitochondrial DNA analysis tested whether these morphology-based groups reflected actual evolutionary lineages, finding substantial discordance.

The traditional classification of monitor lizards has organised the genus Varanus into several subgenera based on anatomical and geographic criteria. These include subgenus Varanus (the nominotypical group, intended to contain the largest and most generalised species), Odatria (small Australasian pygmy monitors), Euprepiosaurus (Indo-Pacific tree and mangrove monitors), and others. Ast's mitochondrial data challenged two of these groupings in ways that have driven taxonomic discussion ever since.

Subgenus Varanus found polyphyletic. The species conventionally placed in the nominotypical subgenus — including members of the gouldii group, V. komodoensis, V. varius, V. salvator, and V. indicus — do not form a single unified clade in Ast's tree. A polyphyletic grouping is one in which the member species are distributed across different parts of the phylogeny, united only by morphological similarity (often convergent) rather than by exclusive shared ancestry. This finding meant that the nominotypical subgenus, as then defined, was an artificial construction that cut across genuine evolutionary lineages.

Subgenus Odatria found paraphyletic. A paraphyletic group is one that contains a common ancestor and some but not all of its descendants — it is a partial, not a complete, evolutionary unit. Ast's analysis found Odatria to be paraphyletic, with the gouldii group of large Australian monitors nested within or adjacent to what had been classified as the dwarf-monitor subgenus. This implies that the boundary between "small Australasian monitors" and "large Australasian monitors" does not correspond to a clean split in ancestry.

These results did not invalidate the concept of subgenera in Varanus — they called for their redefinition on phylogenetic rather than purely morphological grounds. In the decades following Ast (2001), further molecular studies using additional nuclear and mitochondrial markers have largely confirmed the three-clade structure while adding resolution within each group. The classification of V. komodoensis at the species level has never been in doubt; it is the arrangement of species groups and subgenera around it that the molecular era has continued to revise.

Molecular Versus Fossil Evidence

The dialogue between molecular phylogenetics and the fossil record is one of the most productive tensions in varanid biology. Molecular studies reconstruct evolutionary history from DNA sequences of living species, inferring divergence times through molecular clock calibrations; the fossil record provides direct physical evidence of past species and their morphology, but is fragmentary and geographically uneven across the monitor lizard range.

One of the most productive tensions in varanid biology is the dialogue between molecular phylogenies and the fossil record. Ast's mtDNA study operates almost entirely in the present: it samples living species, sequences their DNA, and infers their evolutionary history from patterns of nucleotide similarity. The fossil record, by contrast, provides direct evidence of ancient forms but suffers from incompleteness and the difficulty of placing fragmentary specimens on trees derived from soft-tissue characters such as gene sequences.

In broad terms, the three-clade topology that Ast recovered — African basal, Indo-Asian intermediate, Indo-Australian derived — is compatible with a scenario in which Varanus originated in Africa or the western Tethys, dispersed eastward through Asia, and eventually colonised Australasia. This directional dispersal model fits a reasonable reading of the fossil record, which shows the genus in Europe and Asia by the Miocene and in Australia by the mid-Cenozoic. However, the molecular data do not provide dates; the rate at which mitochondrial genes accumulate mutations varies across lineages and must be calibrated against fossil-derived divergence times, introducing uncertainty.

A specific tension concerns the gouldii group and its relationship to Odatria. Morphological analyses had generally separated large and small Australian monitors cleanly, placing them in distinct subgenera. The molecular paraphyly of Odatria implies that body-size reduction may have occurred multiple times within the Australian radiation, rather than once. This kind of homoplasy (independent evolution of similar traits) is common in lizards and does not undermine the phylogeny, but it does mean that ecological and morphological similarity is a less reliable guide to kinship than molecular sequence data.

For readers interested in how the fossil record of Megalania interacts with the molecular placement of V. komodoensis, the detailed treatment on the fossil record and Megalania page is the appropriate companion. The key take-home from Ast's work is that the molecular data independently supported an Australian ancestry for the Komodo dragon well before the palaeontological evidence from Timor and Queensland was fully assembled.

Myths vs Facts

Common Assumption (pre-2001)	What Ast's mtDNA Evidence Shows
The Komodo dragon is most closely related to large Asian monitors such as V. salvator.	Molecular data place V. komodoensis in the Indo-Australian clade, with V. varius of eastern Australia as its closest living relative.
Subgenus Varanus (s.s.) is a natural evolutionary unit defined by shared ancestry.	The nominotypical subgenus is polyphyletic — its members are scattered across different branches — making it an artificial grouping under phylogenetic species concepts.
Small Australian pygmy monitors (Odatria) form a single, self-contained clade.	Odatria is paraphyletic; the large Australian gouldii-group monitors are embedded within or adjacent to it, blurring the large/small divide.
The giant size of V. komodoensis evolved in isolation on Flores, driven by island conditions or prey availability.	Ast's phylogeny is consistent with large body size being ancestral to the komodoensis–varius clade, predating the colonisation of the Lesser Sundas.
Molecular and morphological phylogenies of monitors broadly agree at all levels.	The mtDNA topology conflicts with several morphologically based subgenus arrangements, particularly regarding branching order among major groups.
African monitors are derived members of the Varanus radiation.	The parsimony tree places the African clade as sister to all other Varanus — the most basally divergent group, not a derived one.

Key Takeaways

• Three-clade structure confirmed. Ast (2001) established African, Indo-Asian, and Indo-Australian groupings within Varanus that have been reinforced by subsequent molecular studies and are now the accepted framework for monitor phylogenetics.
• Komodo dragon belongs to an Australian radiation. The mitochondrial data place V. komodoensis firmly in the Indo-Australian clade, with V. varius as its closest living relative — contradicting prior assumptions of Asian affinity.
• Traditional subgenera required revision. The nominotypical subgenus Varanus is polyphyletic and Odatria is paraphyletic under the molecular tree, necessitating reconsideration of long-standing classification schemes.
• Molecular and fossil evidence converge on Australian origin. The Ast topology predicted an Australian ancestry for V. komodoensis that palaeontological discoveries from Queensland and Timor have since independently corroborated (see Hocknull et al. 2009).
• Varanoidea monophyly confirmed. Outgroup analysis using Heloderma and Lanthanotus confirmed that all three varanoid lineages descend from a single common ancestor, with Lanthanotus as the closest relative of Varanus.
• A single most parsimonious tree recovered. The clean outcome of the parsimony analysis gave Ast's topology an unusual degree of resolution for a molecular study of its era, lending authority to findings that were contested at the time.

Frequently Asked Questions

Who is Jennifer C. Ast and where was this research conducted?

Jennifer C. Ast conducted this research at the Museum of Zoology, University of Michigan, Ann Arbor. The paper was published in the journal Cladistics in 2001 (volume 17, pages 211–226). The Museum of Zoology at Michigan has a long tradition of systematic and evolutionary biology research and maintains extensive comparative collections that support phylogenetic studies of this kind.

What exactly is a mitochondrial phylogeny and why use it for monitors?

A mitochondrial phylogeny is an evolutionary tree built from DNA sequences of genes located in the mitochondrion — a cellular organelle with its own small genome distinct from the nuclear chromosomes. Because mitochondrial DNA is maternally inherited without recombination, it accumulates mutations in a relatively orderly fashion, making it useful for inferring evolutionary relationships. By the late 1990s, mitochondrial gene sequencing was a mature, cost-effective method that made broad taxonomic surveys like Ast's feasible across dozens of Varanus species.

What does it mean that subgenus Varanus is polyphyletic?

Polyphyly means that the members of a named group are distributed across separate, unrelated branches of the evolutionary tree. For subgenus Varanus, this means that species traditionally assigned to it — including V. komodoensis, V. varius, V. gouldii, and others — do not share a most recent common ancestor that is exclusive to the group. They resemble each other in certain ways, but those resemblances are convergent rather than inherited from a single common ancestor. Polyphyletic groups are rejected in modern phylogenetic taxonomy because they misrepresent evolutionary history.

Is V. varius (the lace monitor) really more closely related to the Komodo dragon than V. salvator (the water monitor) is?

That is what Ast's mitochondrial data indicate, and the finding has been supported by subsequent molecular studies. Despite the fact that V. salvator inhabits the same broad Indo-Pacific region as V. komodoensis and overlaps with it geographically on some Indonesian islands, geographic proximity is not a reliable guide to phylogenetic closeness. The molecular evidence consistently places V. komodoensis and V. varius together in the Indo-Australian clade, while V. salvator belongs to the separate Indo-Asian lineage.

How does this paper relate to the Megalania question?

Ast (2001) did not include extinct taxa such as Varanus priscus (Megalania) directly — fossil specimens cannot provide DNA under current methods for material of that age. However, the phylogenetic framework Ast established placed V. komodoensis and V. varius as sister taxa, and subsequent researchers used this topology to argue that large body size may be ancestral to this lineage, predating the Komodo dragon's arrival in the Lesser Sundas. The palaeontological evidence on Megalania and its relationship to the Komodo dragon is explored in the fossil record page and the Hocknull et al. (2009) summary.

Were Ast's results contested after publication?

Like most molecular phylogenies of the era, Ast's results entered a field where morphological systematists sometimes questioned the reliability of single-gene or mitochondrial-only analyses. Mitochondrial DNA can be subject to issues such as numts (nuclear copies of mitochondrial sequences) and rate variation across lineages. Subsequent studies incorporating nuclear markers and additional taxa have largely confirmed the three-clade framework and the placement of V. komodoensis in the Indo-Australian clade, lending broad support to Ast's core conclusions while adding resolution to disputed internal nodes.

Does Varanoidea monophyly have implications for the venom question?

Yes. The confirmed monophyly of Varanoidea — with Heloderma (the venomous Gila monster) as an outgroup and Lanthanotus as the closest relative of Varanus — is relevant to the question of whether varanid venom is an ancestral trait of the group. If venom glands are homologous across Varanoidea, their presence across monitor species would be consistent with inheritance from a common ancestor. This question is addressed in detail in the Fry et al. (2009) paper review.

How has the field of varanid molecular phylogenetics moved on since 2001?

Considerably. The years following Ast (2001) saw the introduction of additional mitochondrial loci, nuclear genes, and eventually genome-wide datasets. Methods shifted from parsimony to model-based approaches such as maximum likelihood and Bayesian inference, which handle rate variation among sites more explicitly. Taxon sampling expanded to include newly described species — the number of recognised Varanus species has grown substantially since 2001. These analyses have generally validated the three-clade topology while resolving previously ambiguous relationships within each major group. Ast's study thus occupies a historically important position: it was not the final word, but it was among the most rigorous opening arguments in the molecular-phylogenetic era of varanid systematics.

Sources & Further Reading

1. Ast, J.C. (2001). "Mitochondrial DNA evidence and evolution in Varanoidea (Squamata)." Cladistics 17: 211–226. https://doi.org/10.1111/j.1096-0031.2001.tb00118.x — Primary source for this summary.
2. Fuller, S., Baverstock, P. & King, D. (1998). "Biogeographic origins of goannas (Varanidae): a molecular perspective." Molecular Phylogenetics and Evolution 9(2): 294–307. An earlier molecular study suggesting close affinity between V. komodoensis and V. salvadorii; contrasts with Ast's sister-taxon finding.
3. Pianka, E.R., King, D.R. & King, R.A. (eds.) (2004). Varanoid Lizards of the World. Indiana University Press. Comprehensive reference on monitor biology, taxonomy, and natural history that incorporates the Ast (2001) molecular framework.
4. Hocknull, S.A. et al. (2009). "Dragon's paradise lost: palaeobiogeography, evolution and extinction of the largest-ever terrestrial lizards (Varanidae)." PLOS ONE 4(9): e7241. https://doi.org/10.1371/journal.pone.0007241 — Fossil evidence converging on Ast's Australian-origin conclusion; see also our editorial summary.
5. Fry, B.G. et al. (2009). "A central role for venom in predation by Varanus komodoensis (Komodo dragon) and the extinct giant Varanus (Megalania) priscus." Proceedings of the National Academy of Sciences 106(22): 8969–8974. https://doi.org/10.1073/pnas.0810883106 — Uses varanoid phylogeny (including Ast's framework) to contextualise the venom discovery.
6. Vidal, N. & Hedges, S.B. (2005). "The phylogeny of squamate reptiles (lizards, snakes, and amphisbaenians) inferred from nine nuclear protein-coding genes." Comptes Rendus Biologies 328(10–11): 1000–1008. Post-Ast nuclear-gene study relevant to broader squamate relationships including Varanoidea.
7. Collar, D.C. et al. (2011). "Biting disrupts integration to spur skull evolution in ecoevolving lacertid lizards." Evolution 65: 1312–1330. Example of how Ast's phylogenetic framework was used in comparative morphological analyses following publication.