Coloration, caudal-peduncle weapons, and venom in Acanthuridae: a critical review

Toward a phylogenetic comparative program

Author

M. E. Alfaro (compiled with research assistance)

Published

August 27, 2026

1 Executive summary

Surgeonfishes (Acanthuridae) share a family-defining feature — one or more sharp, modified scales on the caudal peduncle — that has anchored their common name, their taxonomy, and their popular reputation as reef combatants for nearly two centuries. Yet the empirical literature on how these caudal weapons function, whether they are integrated with coloration into a coordinated signal system, and whether any acanthurid caudal structure delivers venom is much thinner than the volume of secondary claims would suggest. Five conclusions rise consistently out of the primary literature: (1) all extant acanthurids possess derived, modified caudal-peduncle armament, but the armament has evolved along two structurally distinct lines — the erectile, socketed “scalpel” of Acanthurinae (single per side in Acanthurus, Ctenochaetus, Zebrasoma, Paracanthurus, multiple in Prionurus) and the fixed, keel-like bony plates of Naso — with a coherent osteological, myological, and molecular phylogenetic backbone documenting the transitions [@winterbottom1971; @winterbottom1993; @winterbottommclennan1993; @klanten2004; @sorenson2013]; (2) direct behavioural evidence for caudal-spine deployment in intraspecific and interspecific combat is limited but real, and is best-documented in Acanthurus leucosternon [@schober1992]; (3) territoriality and interference competition are widespread but far from universal within the family, and are strongest in a subset of Acanthurus and Ctenochaetus-like species that also show conspicuous flank or caudal coloration [@robertson1986; @craig1996]; (4) sexual dimorphism in weapon-adjacent morphology has been formally documented in Naso unicornis, but has not been reported for scalpel size or shape in any Acanthurus [@demartini2016]; and (5) there is no confirmed evidence that the caudal scalpel of any acanthurid delivers venom via a specialised gland or grooved delivery structure — the well-known Smith & Wheeler [@smith2006] phylogenetic road-map of piscine venoms places Acanthuridae outside the twelve venomous fish clades that they identified anatomically.

The most promising research directions are: rigorous receiver-perspective visual modelling of localised caudal contrast (rather than whole-body colour diversity), because the interesting evolutionary variation is at the peduncle, not the whole animal; comparative biomechanics of weapon geometry (tip radius, curvature, second moment of area, deployment kinematics) across all six extant genera; and a definitive anatomical + transcriptomic audit of caudal integumentary tissue in Acanthurus, Prionurus, and Naso, so that the venom question can be settled at the level of primary evidence rather than by inference from painful human wounds.

2 Search methodology

Databases and platforms. Literature was retrieved between 2026-08-25 and 2026-08-27 using: (a) the scite MCP corpus (>210 M records; full-text excerpts and Smart Citation snippets); (b) targeted Web search restricted to publisher pages, Crossref, PubMed, and institutional repositories; and (c) direct DOI verification via WebFetch. All references retained in the bibliography were verified against at least one authoritative source and are catalogued in the appendix.

Inclusion criteria. Peer-reviewed primary literature and monographs. Aquarium websites, hobbyist blogs, and Wikipedia were used only as pointers toward the underlying paper. Preprints were included only where citation was to a specific dated version and where the empirical content bore directly on a research gap.

Date range. All years accepted (earliest cited work: Winterbottom [@winterbottom1971]; most recent cited work: Zheng et al. [@zheng2026] preprint). No sunset filter was applied because the core anatomical papers on acanthurid weapons are from the 1970s and 1990s and remain the primary sources.

Evidence grading. Each empirical claim is tagged inline where necessary: direct experimental (behavioural or biomechanical test with controls); direct observational (systematic behavioural observation); anatomical/histological; biochemical/functional (assay of secretion or extract); comparative/phylogenetic (correlated evolution across a tree); anecdotal (single observation or aquarium report); unsupported assertion (repeated in reviews without a primary citation).

Limitations. Two central primary sources — Schober & Ditrich (1992) on Acanthurus leucosternon [@schober1992] and Randall’s monograph Surgeonfishes of the World [@randall2001] — were not open-access at the time of retrieval; the former was verified via its DOI record and via multiple citing papers that quote its findings [@bellwood2017scopas; @zheng2026], and the latter is a widely cited monograph whose full contents could not be independently inspected in this review. Where I rely on Randall’s textbook I say so.

3 Taxonomic and anatomical overview

Acanthuridae as currently understood comprise six extant genera and roughly 85 valid species [@ludt2019; @sorenson2013]. Molecular and morphological phylogenies concur that Naso (Nasinae) is sister to a well-supported Acanthurinae containing Prionurus, Paracanthurus, Zebrasoma, Ctenochaetus, and Acanthurus [@winterbottom1993; @winterbottommclennan1993; @klanten2004; @sorenson2013]. Within Acanthurinae, Prionurus is basal, Paracanthurus + Zebrasoma form one clade, and Acanthurus + Ctenochaetus form another with Ctenochaetus nested inside a paraphyletic Acanthurus [@sorenson2013; @lungstrom2025]. The family is placed in Acanthuriformes with Zanclus (Zanclidae) and Luvarus (Luvaridae) as successive out-groups; †Kushlukiidae and several Bolca acanthuriform fossils extend the crown to at least the early Eocene [@bannikov1995; @carnevale2024].

Two structurally distinct weapon systems. All Acanthuridae possess modified caudal-peduncle scales that are highly derived relative to the flat cycloid or ctenoid scales of most percomorphs. But the derived states themselves take two very different forms:

In Acanthurinae (Prionurus, Paracanthurus, Zebrasoma, Ctenochaetus, Acanthurus) the modified scales are erectile “scalpels”: a curved, laterally compressed, sharp-edged modified scale that lies flat in a groove or socket along the peduncle when the fish is at rest, and swings outward (rotates on its base) when the fish arches or flexes its caudal region. Winterbottom [@winterbottom1971] provided the primary anatomical and functional description of this movement mechanism, showing that lateral rotation of the spine is driven by extrinsic axial and hypaxial musculature and that the folding-spine condition — in which the spine is held closely appressed against the peduncle at rest but can be presented at ~90° during aggressive posturing — is the derived condition inside the family. Guiasu & Winterbottom (as summarised by [@bannikov1995]) and Winterbottom [@winterbottom1993] provided the osteological synapomorphies that support this reconstruction. Most Acanthurinae have a single scalpel on each side of the peduncle; Prionurus uniquely bears multiple keeled scutes (usually three to ten fixed, keeled plates in tandem) on each side of the peduncle [@ludt2019]. Zebrasoma scopas shows within-species variation in whether the scalpel is smooth, sharp, and retractable versus bears forward-directed spikes near its base [@bellwood2017scopas], the first documented intraspecific morphotype of the acanthurine scalpel.
In Nasinae (Naso) the caudal armament is composed of one or two fixed, shield-shaped bony plates on each side of the peduncle (a single fixed plate in the subgenus Axinurus, two plates in all other Naso) [@zheng2026]. These are not erectile, and they project forward as sharp keels that engage the water and any opposing tissue when the caudal region is swept laterally. In several Naso species the plates are large enough to give the peduncle a diamond-shaped cross-section; the anterior plate typically bears more wear than the posterior plate, indicating asymmetric contact with substrate or opponents [@demartini2016]. A single case of unilateral absence of both plates in an adult male Naso tergus [@zheng2026] shows that the plates are not indispensable for survival.

Ontogeny of the weapon. The scalpel or plate does not appear at settlement. Juvenile Acanthuridae pass through a pelagic “acronurus” larval stage; the earliest Eocene fossils that can be assigned to the scalpel-bearing clade are already at that stage [@tyler2011glarithurus]. The scalpel emerges and enlarges post-settlement, and in Acanthurus it grows disproportionately relative to body length in many species (positive allometry has been noted informally in the taxonomic literature [@bernal2011]; a proper quantitative allometry of scalpel length vs. standard length across the family has not, to my knowledge, been published).

Anatomy of the “socket.” In Acanthurus the resting scalpel is sheathed within an integumentary groove ringed by a thickened rim of connective tissue; the exposed edge of the scalpel is typically visible externally as a shallow slit. The socket in life is surrounded, in many species, by a highly contrasting patch of pigment (see §4.3). No histological study has yet demonstrated venom-secreting glandular tissue in this sheath in any acanthurid (see §4.4).

Phylogenetic pattern. The multi-plate Prionurus condition is basal within Acanthurinae, and the single-scalpel + folding + retractable state is the derived condition shared by Paracanthurus, Zebrasoma, Ctenochaetus, and Acanthurus [@winterbottom1993]. Within Nasinae the two-plate condition is the plesiomorphic state and the one-plate Axinurus condition is derived. The transition from Nasine bony plates to Acanthurine erectile scalpel is coincident, on the tree, with the emergence of substrate-associated grazing from an ancestor that likely occupied more mid-water niches [@friedman2016; @lungstrom2025]. This does not, on its own, imply a functional causal link; but it means that comparative treatment of weapon evolution must control for feeding-mode and habitat-mode transitions.

4 Critical narrative review

4.1 Morphology and biomechanics

The primary sources on acanthurid caudal-weapon morphology remain Winterbottom [@winterbottom1971; @winterbottom1993] (folding-spine mechanism and skeletal/myological synapomorphies), Guiasu & Winterbottom (osteology of Acanthurinae; see summary in [@bannikov1995]), and Schober & Ditrich [@schober1992] (anatomy plus behavioural deployment of the Acanthurus leucosternon spine). To this small anatomical core, more recent work has added: micro-CT of a A. nigrofuscus deformity showing the peduncle-vertebra articulation [@goatley2018]; the SEM description of intra-specific spike morphotypes in Zebrasoma scopas [@bellwood2017scopas]; the mitochondrial-genome-anchored placement of Prionurus with two multi-plated congeners fully sequenced [@ludt2019]; and a recent phylomorphospace of surgeonfish body and fin shape that treats caudal-spine area as a shape variable and reports its covariation with body elongation and caudal-fin aspect ratio [@lungstrom2025].

Deployment mechanics. Winterbottom [@winterbottom1971] framed the folding scalpel as functionally analogous to a spring-loaded cutting blade: at rest it lies within the socket with only its dorsal edge exposed, and lateral rotation of the scale about its basal attachment is triggered by dorsoventral flexion of the caudal peduncle plus dorsal rotation of the caudal fin. In the deployed position the sharp anterior edge of the scalpel projects outward and forward. Deployment is coupled to the axial swimming musculature and requires no dedicated erector muscle system — the spine “opens” as a consequence of body flexion, and “closes” as the peduncle straightens. This makes the acanthurine scalpel unusually inexpensive to actuate: it is essentially always ready, and the fish reveals it as a side-effect of the same lateral tail-sweeps used to swim, feed, and orient. In Naso the fixed plates have no analogous mechanical closure — they are always engaged, and defence or offence with them is entirely a matter of body posture and speed of lateral tail-sweep [@demartini2016].

Wear as a proxy for use. DeMartini [@demartini2016] reports that in Naso unicornis the anterior peduncular plate shows a higher proportion of visible wear and damage than the posterior plate, and that plate width scales positively with body size and shows sexual dimorphism. He does not quantify wear against behavioural observations of contact. But because wear is one of the few cheap, non-invasive indicators of contact history, this is an under-used variable. There is no equivalent published wear-frequency dataset for Acanthurus or Prionurus.

Biomechanical performance. No published study has measured tip radius, edge curvature, second moment of area, bending stiffness, fracture toughness, or contact force for an acanthurid caudal weapon. The size of this gap is worth emphasising because these are precisely the variables the animal-weapon literature has shown to be most informative about function [@emlen2008; @palaoro2022] — and precisely the variables that are cheapest to measure on museum specimens. Verbal descriptions in the primary literature emphasise sharpness, hardness, and asymmetric bevelling of the edge, but no quantitative micro-CT or nano-indentation dataset exists.

4.2 Behaviour and function

Direct behavioural observation of caudal-scalpel deployment in the field is unusually sparse for such a conspicuous morphological feature. The single most important primary source is Schober & Ditrich [@schober1992], who quantified the anatomy of the scalpel in Acanthurus leucosternon and reported that a pair-bonded individual attacked and cut an intruder using its erected caudal spines during territorial interactions. Their paper is the direct, cited, primary evidence that surgeonfish caudal spines are used offensively in intraspecific aggressive encounters — a claim widely repeated but rarely traced to its source [@bellwood2017scopas; @zheng2026].

Territoriality is heterogeneous. Robertson & Gaines [@robertson1986] examined all thirteen surgeonfish species in a 0.225-ha Aldabra reef assemblage: eleven of these defended feeding territories intraspecifically, most also interspecifically. Species pairs sorted into “interactors” and “non-interactors”; interactors had higher diet overlap and lower large-scale habitat overlap. Dominance was highly asymmetrical within pairs, and reciprocal removal showed that agonistically subordinate species readily invaded vacated habitat of dominant species. This is the most comprehensive comparative-observational dataset on interference competition in the family. It does not report scalpel-deployment counts per encounter, and it is not designed to test whether scalpel morphology predicts dominance rank.

Population-level territoriality. Craig [@craig1996] quantified time-budgets and territory defence in Acanthurus lineatus in American Samoa: >99% daily site fidelity in recognisable individuals, ~1900 daily defence events per fish, ~17000 daily bites, and 60–75% reductions in feeding and defence during rough surf. Compared to Australian and Indian Ocean A. lineatus populations, Samoan fish were smaller-bodied, held smaller territories, and defended at higher rates. This is direct behavioural evidence that territorial Acanthurus invest enormously in signalling and combat activity — but again, the paper does not decompose defence into “displays with scalpel” vs. other display modalities.

Choat & Bellwood [@choat1985] documented that Ctenochaetus striatus — a detritivore, not a herbivore — commonly feeds inside territories defended by A. lineatus, and is tolerated rather than attacked, presumably because dietary overlap is low (see also [@tebbett2017]). This suggests that acanthurid territorial signalling is discriminating, targeting resource competitors and not merely conspecifics.

Schooling and antipredator behaviour. Acanthurus coeruleus and A. bahianus (Caribbean) and several Naso species form large aggregations; both Catano et al. [@catano2014; @catano2015] argue that these species alter foraging behaviour less in response to predator cues than do parrotfishes without a caudal scalpel, and infer that the physical armament may reduce the effective cost of foraging in high-risk microhabitats. This is a correlational, not causal, argument — the scalpel is co-varying with body shape, gut length, feeding kinematics, and social organisation, any of which could explain the risk-tolerance differential.

Deployment during predator interactions. I found no published primary observation of a surgeonfish successfully wounding a predator (e.g. a large piscivore, moray, shark) with its caudal spine. Aquarium and dive lore contain many such claims, but the primary literature does not. This does not falsify the antipredator hypothesis; it does mean the antipredator hypothesis is weakly grounded in direct evidence.

4.3 Coloration

Acanthuridae display extraordinary between-species coloration diversity — mostly bright single-colour bodies with contrasting bars, dots, or stripes; some species (e.g. Zebrasoma flavescens, Paracanthurus hepatus) are among the most instantly recognisable reef fishes. But most of that whole-body diversity is not what the animal-weapon literature would predict is most informative. The evolutionarily interesting variable is localised, receiver-directed contrast around the weapon: how conspicuous the socket, spine, and adjacent flank patch are, both against the fish’s own body and against the reef background.

Peduncular contrast is highly variable. Species like Acanthurus leucosternon, A. lineatus, A. achilles, A. sohal, and A. nigricans have flank and peduncular pigment patterns that produce sharp, high-contrast rings, dots, or arcs immediately around the scalpel socket. Species like most Ctenochaetus have low-contrast peduncular pigmentation. Naso lituratus and N. brevirostris have brightly coloured (yellow, red, orange) peduncular plates that are visibly distinct from the surrounding brown or grey body. Whether these contrasts are aposematic (directed at predators), badge-of-status signals (directed at conspecific rivals), or receiver-independent side-effects of species-recognition selection is unresolved. To my knowledge no published study has performed receiver-specific visual modelling on the peduncular patch in any acanthurid.

Rapid colour change during aggression. Multiple field reports and taxonomic sources describe rapid colour change in acanthurids during aggression, spawning, or stress. Myrberg, Montgomery & Fishelson [@myrberg1988] describe reproductive colour changes and daily spawning behaviour in Acanthurus nigrofuscus, Ctenochaetus striatus, and Zebrasoma xanthurum in the Red Sea. Craig [@craig1996] and Robertson & Gaines [@robertson1986] report peduncular and body colour changes during territorial displays in several Acanthurus. Rapid change specifically in the vicinity of the caudal spine has been observed anecdotally in several Acanthurus (a bright pale patch around the socket becomes momentarily white against a suddenly darker flank) but has not been quantified spectrophotometrically in any published dataset I could locate.

Visual environment. Marshall and colleagues [@marshall2023] argue that the long-wavelength-sensitive opsin (lws) expression, and hence red-sensitivity, of reef fishes covaries with algivory and with the use of orange/red colour signals. Acanthurids were not the direct focus of that paper, but the reasoning applies: whether an acanthurid can “read” a rival’s peduncular signal depends on that rival’s spectral output, on the ambient illumination geometry at the reef microhabitat, and on the visual pigment complement of the receiver. Serially, none of these has been characterised for a system involving a caudal-weapon patch.

Ontogenetic and sex-linked colour changes. Multiple Acanthurus, Zebrasoma, and Naso species show pronounced juvenile-adult colour shifts, and Acanthurus pyroferus famously mimics juvenile Centropyge angelfishes. Sexual dichromatism at spawning has been reported for A. nigrofuscus [@myrberg1988]. But there is no published atlas of ontogenetic peduncular-patch colour change coordinated across species.

Whole-body vs localised colour. A methodological point worth emphasising for the comparative programme is that most published reef-fish colour work quantifies whole-body colour diversity — mean reflectance, primary and secondary hue counts, coverage-weighted colour disparity — and correlates it with clade, habitat, or feeding guild. That framework is well developed but is agnostic to where on the fish the colour sits. In the acanthurid case the question is not “is Acanthurus leucosternon colourful” (it is), but “is the specific patch of pigment around the scalpel socket unusually contrasting against the local flank and the local reef background, and does that contrast scale with weapon investment, territoriality, or sex?” This second question demands localised reflectance sampling and receiver-perspective visual modelling — approaches that are standard in the damselfish [@marshall2023] and cichlid literatures but have not, to my knowledge, been applied to the acanthurid peduncle. The reef-fish visual environment is spectrally structured — outer-reef blue-water and shallow-reef green-water microhabitats have distinct irradiance spectra [@horodysky2013; @marshall2023] — and long-wavelength sensitivity in the receiver is tuned to feeding ecology and to signal use; both matter for the peduncular-signal question.

4.4 Venom — special care

The safest primary-literature statement is: no confirmed histological, biochemical, or genomic evidence exists for a venom-delivery apparatus associated with the caudal scalpel or peduncular plates in any acanthurid. The comprehensive Smith & Wheeler phylogenetic road-map of piscine venoms [@smith2006], which estimated that >1200 fish species should be presumed venomous based on both phylogeny and anatomical inspection of >100 species, identified twelve venomous fish clades — and Acanthuridae were not among them. Sivan’s review of piscine venoms [@sivan2009] and the more recent Frontiers review [@ndandala2023] likewise identify siganids (rabbitfishes, the sister family of Luvaridae + Acanthuridae) as venomous (dorsal, anal, and pelvic fin spines), but treat Acanthuridae as non-venomous. The venom-associated clades in fishes typically pair grooved or channelled dorsal, anal, or opercular spines with cranial glandular tissue [@smith2006; @ndandala2023]; the acanthurid caudal weapon system has neither a grooved spine nor documented glandular tissue.

Where the “venomous surgeonfish” claim originates. The claim is repeated in dive-medicine sources, aquarium websites, and some clinical toxicology reviews. Three phenomena appear to feed it:

Mechanical wounding of humans. The scalpel is genuinely sharp and can inflict deep lacerations; wounds are painful and prone to secondary infection [@sivan2009]. This is trauma, not envenomation.
Toxic flesh / ciguatera. Large Naso, Ctenochaetus, and Acanthurus consumed as food have been implicated in ciguatera poisoning (accumulation of dinoflagellate-derived toxins in flesh). This is a food-web phenomenon and has nothing to do with the caudal weapon.
Extrapolation from siganids. Siganidae, an out-group family sharing a common percomorph ancestor with acanthurids, are venomous [@smith2006]. Sibling-clade venom does not license inference of homology to Acanthuridae.

Anatomical audit. No published histological study has demonstrated glandular tissue in the caudal-peduncle sheath of any Acanthurus, Ctenochaetus, Zebrasoma, Paracanthurus, Prionurus, or Naso. No proteomic or transcriptomic study has recovered a venom-like toxin from acanthurid caudal tissue. The Zheng et al. [@zheng2026] note on an N. tergus individual missing its bony plates makes no claim about venom, and the systematic literature on Naso [@klanten2004; @ludt2019] describes no such structure.

Dorsal and anal fin spines. Acanthurids possess sharp dorsal (typically nine) and anal (typically three) fin spines. These are structurally simple, not grooved, and no histological description of glandular tissue at their base has been published for any acanthurid, to my knowledge. Anecdotally, dorsal-spine wounds are painful in aquarium settings, but again, painful is not venomous.

Bottom line. Until direct histological, biochemical, or genomic evidence appears, the correct scientific statement is that Acanthuridae are not known to be venomous, either at the caudal spine, the peduncular plates, or the median fin spines, and that all “venom” claims traceable to primary literature refer to (a) mechanical wound pain, (b) secondary infection, or (c) dietary ciguatera.

4.5 Sexual dimorphism and ontogeny

Documented dimorphism. DeMartini [@demartini2016] used measurements from Hawaiian Naso unicornis to show that posterior peduncular-plate width, corrected for fork length, is significantly greater in males than in females and can be used as a diagnostic sex predictor. The anterior plate shows more wear than the posterior plate in both sexes but wear patterns did not differ significantly by sex. No comparable dimorphism analysis has been published for scalpel length or curvature in Acanthurus, Ctenochaetus, Zebrasoma, Paracanthurus, or Prionurus.

Body-size dimorphism. Some Naso species show pronounced male-limited elaboration of unrelated structures — the horn or rostral projection of N. brevirostris, N. unicornis, N. annulatus — and males of many Naso attain larger body sizes than females. Taylor et al. [@taylor2014] and Ford et al. [@ford2016] provide age-based demographic evidence that female N. unicornis mature at larger body size than males in Micronesia, but that N. lituratus shows the opposite pattern in Guam. So even in the same genus, size-at-maturity dimorphism can invert between congeners.

Non-sex-limited weapon use. Craig [@craig1996], Robertson & Gaines [@robertson1986], and Choat & Bellwood [@choat1985] all describe territorial Acanthurus lineatus, A. leucosternon, and A. nigrofuscus individuals of both sexes defending feeding territories. This pattern — female weapon use — is a comparative diagnostic that distinguishes the acanthurid case from classical male-only sexually-selected weapon systems and matches Stankowich’s argument [@stankowich2012] that “weapons” that evolve outside a strict sexual-selection context are widely distributed across mammals as well.

Ontogenetic shifts. Naso brevirostris, N. vlamingii, and several other Naso undergo juvenile-to-adult dietary shifts (benthic grazing → gelatinous zooplanktivory) that are associated with body-form changes [@friedman2016]. There is no published dataset on ontogenetic change in scalpel size, socket-patch contrast, or aggressive-behaviour rate in any acanthurid.

Contrast with the mainstream animal-weapon literature. The comparative weapon literature — dominated by beetles, deer, and crustaceans — is structured around three empirical patterns: (i) male-limited elaboration, (ii) positive allometry with body size, and (iii) tight coupling between weapon size and combat outcome [@emlen2008; @palaoro2022]. Acanthuridae partially violate all three. Both sexes bear and use the caudal weapon in territorial Acanthurus [@robertson1986; @craig1996; @schober1992]; only in Naso unicornis has explicit male-biased weapon-adjacent allometry been documented [@demartini2016]; and no primary-literature dataset yet links per-species scalpel size to per-encounter dominance outcome. Stankowich’s work on mammalian defensive weaponry [@stankowich2012; @stankowich2016; @stankowich2014] provides a more useful comparator than the classical sexual-selection literature: he argues that many mammalian “weapons” (spines, quills, plates, noxious secretions) evolve outside a sexual-selection context, are borne by both sexes, and correlate with predator exposure and habitat openness. That framework — weapons as ecological, not exclusively sexual, adaptations — fits the acanthurid pattern much better than Emlen’s classical horn / antler framework, and is the frame under which the acanthurid comparative program should be developed.

4.6 Ecology, feeding, and habitat

The Acanthuridae are ecologically diverse: benthic algal grazers (many Acanthurus, Zebrasoma), sediment/detritivores (Ctenochaetus, some Acanthurus), macroalgal browsers (N. lituratus, N. unicornis), and midwater zooplanktivores (A. thompsoni, A. mata, Paracanthurus hepatus, several Naso) [@choat1985; @friedman2016; @miyake2015; @tebbett2017]. Planktivory has evolved at least four times independently within the family, and each transition is associated with parallel morphological convergence (slender body, smaller adductor mandibulae, high-aspect-ratio pectoral fins) [@friedman2016]. Reef-flat colonisation likely tracks the acquisition of high-aspect-ratio pectoral fins during the Miocene [@bellwood2018reefflat; @sorenson2013].

Territoriality and interference competition are largely restricted to a subset of algal grazers with defensible turf patches [@robertson1986; @craig1996]; roving grazers, schooling zooplanktivores, and Ctenochaetus detritivores show much lower rates of intraspecific aggression. This is consistent with resource-competition predictions and provides the ecological substrate for the concentration of the strongest weapon-use behaviour in the territorial-Acanthurus subclade.

4.7 Phylogenetic history

Sorenson, Santini, Carnevale & Alfaro [@sorenson2013] provided the multi-locus time-calibrated hypothesis on which most subsequent comparative work rests: a nine-locus, 76%-species matrix that recovered Naso as sister to Acanthurinae, Prionurus as sister to (Paracanthurus + Zebrasoma) + (Acanthurus + Ctenochaetus), Ctenochaetus nested within a paraphyletic Acanthurus, and crown Acanthuridae dating to ~40–55 Ma. The recent Lungstrom et al. preprint [@lungstrom2025] used 19 loci and 80 species (97% of extant diversity) to recover a very similar topology with revised branch lengths, and to recover Acanthurus paraphyly with respect to Ctenochaetus.

This tree defines the comparative axis along which weapon and coloration evolution must be studied. Key nodes for our questions are: (i) the Nasinae–Acanthurinae split, which coincides with the transition from fixed keel to erectile scalpel; (ii) the origin of Prionurus, where multiple keeled plates give way to a single scalpel per side; (iii) the origin of the Ctenochaetus + Acanthurus clade, where the strongest concentration of territorial resource-defence sits; and (iv) the multiple, independent transitions to zooplanktivory that repeatedly reduce or de-emphasise the demand on the weapon system [@friedman2016].

5 Species-level evidence table

Values are inserted only where I could locate primary-literature evidence. “Unknown” is used deliberately in preference to congeneric extrapolation. Where an “aggression/defence” observation exists but is not linked directly to the scalpel, that is noted.

Genus, species	Weapon # / type	Fixed / mobile / retractable	Relative size or measurement (available)	Sex or ontogeny difference	Observed aggression/defence use	Associated caudal coloration or change	Venom claim (if any) — anatomical location	Evidence type + strength	Primary citation
Acanthurus leucosternon	1 per side; erectile scalpel	Mobile / retractable	Not quantified	Both sexes territorial	Yes — attacks intruder with erected spine	Contrasting black/white/yellow flank + peduncular ring	None credible; wound painful	Direct observational + anatomical (strong for local claim)	@schober1992
Acanthurus lineatus	1 per side; erectile scalpel	Mobile / retractable	Not quantified	Both sexes territorial	Aggressive defence >1900 events/day but scalpel deployment not counted	Blue/yellow horizontal stripes + orange peduncle patch, spawning colour change	None credible	Direct observational (strong for territoriality; weapon deployment unquantified)	@craig1996
Acanthurus nigrofuscus	1 per side; erectile scalpel	Mobile / retractable	Anatomy characterised via double-tailed deformity CT	Aggregation spawning; sex-linked spawning color change	Aggressive resource defence documented in assemblage	Yellow peduncle-spine pigment	None credible	Direct observational + anatomical	@robertson1986; @myrberg1988; @goatley2018
Acanthurus coeruleus (Caribbean)	1 per side; erectile scalpel	Mobile / retractable	Not quantified	Schooling; low aggression as adult	Reduced predation-risk response inferred vs parrotfish	Uniform blue body, contrasting peduncle spine socket	None credible	Comparative/observational (weak for scalpel-specific claim)	@catano2014
Acanthurus bahianus / A. tractus	1 per side; erectile scalpel	Mobile / retractable	Not quantified	Not reported	Reduced foraging suppression under predation risk	Yellow/blue caudal-fin margin (tractus vs bahianus)	None credible	Correlational + colour taxonomic	@bernal2011; @catano2015
Ctenochaetus striatus	1 per side; erectile scalpel	Mobile / retractable	Not quantified	Not reported	Feeds within A. lineatus territories, low aggression	Low-contrast peduncular pigment	None credible	Direct behavioural / dietary	@choat1985; @tebbett2017; @krone2008
Zebrasoma scopas	1 per side; erectile scalpel	Mobile / retractable — with intraspecific variation	Two morphotypes (smooth vs forward-spiked) documented via SEM	Not reported	Not documented	Uniform dark body, contrasting peduncle ring	None credible	Direct anatomical (single-population dataset)	@bellwood2017scopas
Zebrasoma flavescens	1 per side; erectile scalpel	Mobile / retractable	Not quantified	Genomic evidence of local adaptation to habitat	Not documented	Bright yellow body, contrasting peduncular white socket	None credible	Genomic + observational	@bernardi2018
Paracanthurus hepatus	1 per side; erectile scalpel	Mobile / retractable	Not quantified	Not reported	Not documented	Body: black-blue-yellow; peduncular spine yellow socket	None credible	Anatomical + behavioural anecdote	@friedman2016
Prionurus laticlavius / P. biafraensis	Multiple (3–10 per side); fixed keeled plates	Fixed	Not quantified in vivo	Not reported	Aggregating; observed use unclear	Body drab; peduncular plates yellow-cream in some spp.	None credible	Anatomical + phylogenetic	@ludt2019
Naso unicornis	2 plates per side; fixed shield keels	Fixed	Posterior plate width dimorphic; anterior plate wear disproportionate	Yes — sexual dimorphism in posterior plate width	Aggregation forming; combat use anecdotal	Body drab; plate contrast weak	None credible	Direct anatomical + morphometric	@demartini2016
Naso lituratus	2 plates per side; fixed shield keels	Fixed	Not quantified	Sex-difference in maturation size	Territorial behaviour weak	Bright orange plate + yellow face and dorsal fin	None credible	Anatomical + demographic	@taylor2014; @ford2016
Naso brevirostris	2 plates per side; fixed shield keels	Fixed	Not quantified	Male rostral horn; sex differences in size at maturity in some populations	Not documented	Body drab; plate contrast weak	None credible	Anatomical + comparative	@klanten2004; @friedman2016
Naso tergus	2 plates per side (standard); unilateral absence documented	Fixed	412 mm SL exceeds published max — first anomaly record	Not reported	Not documented	Body bright green dorsally	None credible	Anatomical anomaly report	@zheng2026
Naso subgenus Axinurus	1 plate per side (derived)	Fixed	Not quantified	Not reported	Not documented	Unknown	None credible	Systematic	@zheng2026

6 Venom audit table

Every acanthurid species for which I could locate a direct anatomical or biochemical statement bearing on venom is listed. All entries were negative — no verified venom apparatus has been identified in any acanthurid.

Species	Tissue / structure examined	Methods used	Result	Strength of inference
Acanthuridae (family-level survey)	Median-fin spines and caudal peduncle, comparative anatomical inspection	Broad anatomical/phylogenetic road-map	No venom apparatus identified; family excluded from 12 venomous fish clades	Strong — explicit exclusion at family level
Acanthuridae (family-level review)	Fin spines and caudal-peduncle armament	Literature synthesis of venom apparatus	Not identified as venomous	Moderate (secondary synthesis)
Acanthurus leucosternon	Caudal-spine and integumentary sheath	Gross anatomy + behavioural observation	Sharp, cutting scalpel; no gland described	Anatomical — no positive venom finding
Naso unicornis	Peduncular plates	Gross anatomy + morphometry + wear scoring	Fixed keel; no gland described	Anatomical — no positive venom finding
Acanthurus nigrofuscus	Caudal skeleton + spine	Micro-CT of deformed individual	Skeletal and spine anatomy imaged; no gland identified	Anatomical (deformity study)
Zebrasoma scopas	Caudal spine surface	SEM	Smooth vs spiked morphs; no gland/groove reported	Anatomical (surface only)
Prionurus biafraensis, P. laticlavius	Mitochondrial genome	Whole mitogenome	Standard percomorph gene content; no toxin-relevant loci identified	Molecular (not a venom test)
Median-fin spines, various	Sharp dorsal (~IX) and anal (~III) spines	Descriptive taxonomy	Structurally simple, non-grooved; painful mechanical wounds	Anatomical (no gland described)

Diagnostic contrast. In Siganidae, the sister family via Luvaridae [@bannikov1995; @smith2006], grooved dorsal, anal, and pelvic-fin spines are paired with basal glandular tissue that has been directly imaged and biochemically characterised; siganids appear in the venomous-fish road-map [@smith2006]. Acanthuridae do not.

7 Hypothesis-and-prediction table

For each evolutionary hypothesis I lay out the causal logic, the current status of support, and measurable comparative predictions.

Hypothesis	Causal logic	Current status	Sex / ontogeny prediction	Ecological prediction	Discriminating observation / experiment
H1. Resource-competition	Weapon evolves under selection for defensible feeding territories; larger/sharper weapons yield higher hold-time on food patches	Supported by concentration of strongest weapon-deployment behaviour in territorial Acanthurus [@robertson1986; @craig1996]; @schober1992 provides direct causal link in A. leucosternon	Both sexes bear and use weapon; ontogenetic increase co-occurs with acquisition of feeding territory	Weapon-elaboration correlates with defensible-turf habitat (reef flat, wave-swept) and reduced dietary overlap with schooling competitors	Correlation of relative weapon size with territory-holding index across species, controlling for body size + habitat
H2. Predator-defence	Erectile scalpel reduces attack success by piscivores; presence deters or wounds attackers	Weakly supported: schooling Acanthurus show reduced foraging suppression under predator cues [@catano2014; @catano2015], but no direct wounding observation	Both sexes; scalpel appears at settlement or shortly after	Weapon-elaboration correlates with predator exposure in open habitats and diurnal activity	Field predator-experiment; scalpel presence/absence manipulation (ethically challenging); mechanical wound testing on predator dermis
H3. Sexual-selection	Weapon enlarged and used in male-male combat over females; positive allometry expected	Only formally supported for Naso unicornis posterior-plate width [@demartini2016]; no Acanthurus scalpel dataset	Male-biased; positive allometry; enlarges at male sexual maturity	Weapon-elaboration correlates with lek/pair-spawning mating system in Naso	Sex-specific allometric analysis of weapon dimensions across the family; behavioural staging of male–male encounters
H4. Signal–weapon coevolution	Localised caudal contrast evolves to advertise weapon presence to rivals (badge) or predators (aposematism)	Untested — no receiver-perspective visual model exists for any acanthurid peduncular patch	Both sexes; ontogeny tracks weapon emergence	Localised contrast correlates with weapon elaboration and territoriality after controlling for whole-body colour diversity	Reflectance spectrometry + visual modelling of peduncular patches; comparative test of contrast × weapon × territoriality across the tree
H5. Concealed-weapon	Sheathed, cryptic scalpel exploits surprise; conspicuousness should be low around weapon	Predicts opposite pattern to H4; direct test would use inconspicuous vs conspicuous species and encounter kinematics	Both sexes	Cryptic peduncle correlates with ambush-adjacent lifestyle and reduced signalling	Contrast × combat-outcome analysis; kinematic reconstruction of attack posture and receiver line-of-sight
H6. Venom-substitution	Sharp mechanical weapon evolves as substitute for lost or absent venom	Family lacks venom [@smith2006]; if H6 were correct, absence of venom would predict elaboration of mechanical armament. But this pattern is generic to many non-venomous fish.	No specific sex/ontogeny prediction	Weapon-elaboration correlates negatively with any residual venom capacity across acanthuriforms	Comparative venom-audit + weapon-elaboration across Acanthuroidei
H7. Venom-potentiation	Weapon delivers a venom or toxin, either from spine or from adjacent gland	Currently unsupported. No histological or biochemical evidence in any acanthurid [@smith2006; @sivan2009]	N/A until basic evidence is established	If true, wound severity would exceed mechanical expectation; toxin would be detectable via proteomics	Histology and transcriptomics of peduncular sheath in Acanthurus, Prionurus, Naso
H8. Modularity	Weapon, colour, and behaviour are quasi-independent modules that evolve at different rates and in response to different pressures	Partially supported: repeated planktivory transitions decouple weapon from body form [@friedman2016]; peduncular colour appears to vary among close relatives without commensurate weapon change	Different sex/onto patterns across modules	Different ecological correlates for each module	Multivariate phylogenetic path analysis; module-specific evolutionary-rate estimation
H9. Exaptation / co-option	Ancestrally-defensive structure is co-opted for signalling, or vice versa	Compatible with Winterbottom’s ancestral-state reconstruction that the folding condition is derived from a fixed-plate ancestor [@winterbottom1971; @winterbottom1993]	Depends on the trajectory	Depends on the trajectory	Ancestral-state reconstruction of weapon mobility × territoriality; hidden-state model of transitions

8 Proposed comparative dataset

For each species with a specimen series >5 individuals (museum + fresh) I recommend collecting:

Weapon morphometry - Scalpel length, base width, tip radius, edge curvature (2D outline) - Scalpel second moment of area at mid-length (from micro-CT or from lateral scan) - Deployment angle at maximal erection (imaged from euthanised specimens) - Number of scalpels/plates per side (integer; distinguishes Prionurus) - Presence/absence of forward spikes (state; captured on Z. scopas) - Sex-specific allometry (log-scalpel-length ~ log-standard-length with sex interaction)

Body-size and body-shape covariates - Standard length, body depth, caudal-peduncle depth and width - Caudal-fin aspect ratio (already tabulated in [@lungstrom2025]) - Pectoral-fin aspect ratio (already tabulated in [@friedman2016; @lungstrom2025])

Material properties - Bending stiffness and failure load at deployed angle (three-point bend) - Vickers hardness of dentine-like enameloid cutting edge - Fracture toughness of edge under wet loading

Behaviour - Territoriality index (0 = purely roving, 1 = year-round site fidelity) - Mating-system score (broadcast spawning, pair-spawning, lek-like aggregation) - Feeding guild (grazer, detritivore, browser, planktivore) - Habitat openness index (from published habitat use records) - Predator exposure index (from local piscivore densities)

Coloration (localised) - Reflectance spectra at peduncular socket, adjacent flank, and reference dorsal region, under standardised broadband illumination - Michelson contrast between socket and adjacent flank, under water-column illuminants at 3 m depth - Just-noticeable-difference (JND) between socket and body, under receiver-specific visual models for (a) a conspecific with acanthurid-typical LWS/MWS/SWS cone complement, (b) a serranid predator, (c) a labrid predator - Presence/absence of dynamic peduncular colour change during aggression (from video) - Ontogenetic colour trajectory (juvenile → adult samples)

Venom-relevant tissue (for the subset targeted for the venom audit) - Histology of caudal-sheath integument (H&E + Mallory’s trichrome + PAS) - RNA-seq of peduncular integument and comparative control tissue (dorsal flank) - Proteomic screen of macerated sheath tissue against known fish venom-toxin catalogues

9 Analytical recommendations

Phylogenetic comparative approaches. Use the Sorenson et al. [@sorenson2013] tree, with the Lungstrom et al. [@lungstrom2025] preprint topology as sensitivity, as the working phylogenetic hypothesis. Prune to species with morphology + behaviour data; refit branch lengths under a relaxed clock if calibrations change.

Recommended analyses, roughly in order of increasing structural risk:

Multivariate phylogenetic PCA on weapon + body-shape variables (mvMORPH). This handles the well-documented covariation between caudal-spine area and body/caudal-fin shape [@lungstrom2025].
Phylogenetic MANOVA / PGLS of localised peduncular contrast vs weapon size, territoriality, and habitat, controlling for whole-body colour diversity. This is the direct test of whether “localised caudal contrast” is more informative than whole-body colour diversity, per the pre-registered question in the review brief.
Correlated-transition (Pagel) or hidden-state (HiSSE / MuHiSSE) models on discrete weapon states (single scalpel vs multiple keeled plates vs fixed shield plates), fixed vs mobile deployment, territoriality state, and mating system.
Ornstein–Uhlenbeck modelling of weapon-length allometry with regime shifts placed at planktivory transitions [@friedman2016] and at the Nasinae–Acanthurinae split.
Phylogenetic path analysis (phylopath) to compare directed acyclic graphs corresponding to H1–H9. This is the cleanest way to arbitrate resource-competition, sexual-selection, and signal–weapon coevolution hypotheses simultaneously.
Sex-specific evolutionary allometry where sex-matched morphometric data are available. The DeMartini [@demartini2016] N. unicornis dataset is the seed; extension across Naso and across territorial Acanthurus is the priority.

Confounders and limits. Body-form evolution in Acanthuridae is strongly linked to diet (planktivory vs grazing) and locomotor mode (labriform pectoral vs caudal-fin swimmer) [@friedman2016; @lungstrom2025]. Any test of weapon–colour coevolution has to residualise weapon size, tip curvature, and peduncular contrast against body shape and against locomotor mode. Second, the strongest observational behavioural data are all from a small handful of species (A. lineatus, A. leucosternon, A. nigrofuscus, C. striatus, N. unicornis, N. lituratus); comparative claims that generalise beyond those species are formally underdetermined. Third, several Nasinae undergo dramatic ontogenetic shifts in ecology [@friedman2016]; adult-only sampling will misrepresent the selective regime under which weapon-adjacent traits emerged.

10 Research agenda

Priority 1 — Museum measurements. Digitise the entire acanthurid caudal peduncle, in high-resolution lateral view, from every genus-species-sex combination available in the major North American, European, Australian, and Japanese collections. Extract scalpel length, base width, tip radius, edge curvature, and second moment of area. Deposit the dataset publicly. This is the single highest-value low-cost intervention.

Priority 2 — Histology and transcriptomics of the peduncular sheath. Fresh-tissue histology (H&E + PAS) of the caudal integument in Acanthurus leucosternon, A. lineatus, A. nigrofuscus, Ctenochaetus striatus, Zebrasoma scopas, Paracanthurus hepatus, Prionurus laticlavius, and Naso unicornis, plus RNA-seq of the same tissue against a control (dorsal flank) tissue block. This resolves the venom question at the level of primary evidence.

Priority 3 — Mechanical testing. Three-point bending, cutting force under simulated dermal loading, and Vickers hardness of the edge, for at least one representative species per genus.

Priority 4 — Behavioural observation. Filmed, individually-marked territorial encounters in three species (A. lineatus, A. leucosternon, A. nigrofuscus) with per-encounter counts of scalpel deployment vs display alone vs colour change. This is the empirical basis on which every comparative claim ultimately rests.

Priority 5 — Visual modelling. Reflectance spectrometry of peduncular patches for as many species as possible, plus published visual-pigment complements of a small set of receivers (conspecific, serranid, labrid, sphyraenid piscivores), driving JND-based contrast estimates. This resolves the “conspicuousness of the weapon patch” question.

Priority 6 — Venom assays. Only if histology or transcriptomics returns a positive gland/toxin signal in Priority 2; otherwise Priority 6 is unnecessary and the venom question is answered.

11 Concluding assessment

Five conclusions currently well supported.

Acanthuridae uniformly possess derived caudal-peduncle armament, but the armament has evolved along two structurally distinct routes — erectile scalpel in Acanthurinae, fixed keeled plates in Nasinae — with strong osteological, myological, and molecular support for the transition [@winterbottom1971; @winterbottom1993; @winterbottommclennan1993; @sorenson2013].
The erectile scalpel of Acanthurus leucosternon has been directly observed to be used offensively during territorial combat with conspecific intruders [@schober1992]. This is the strongest, most cleanly documented instance of weapon deployment in the family.
Territoriality and interference competition are strongly heterogeneous across the family, concentrated in a subset of algal-grazing Acanthurus and Ctenochaetus-like species that overlap in diet and defend spatially-restricted turf patches [@robertson1986; @craig1996; @choat1985].
Sexual dimorphism in weapon-adjacent morphology has been formally documented in Naso unicornis (posterior peduncular-plate width) [@demartini2016]; no comparable sex-specific dimorphism has yet been established for the erectile scalpel of Acanthurus.
Acanthuridae are not known to be venomous. The comprehensive phylogenetic road-map of piscine venoms places them outside the twelve venomous fish clades [@smith2006]. No histological, biochemical, or genomic evidence for a venom-delivery apparatus in the caudal or median-fin spines of any acanthurid has been published.

Five important but unresolved questions.

Does relative scalpel size (or plate size) scale with positive allometry, and does allometric exponent differ between sexes in any Acanthurus — as it has been shown to in Naso unicornis [@demartini2016]?
Is “localised caudal contrast” (peduncular socket vs adjacent flank, under receiver-specific visual models) more informative about territoriality and weapon presence than whole-body colour diversity?
What is the histology and transcriptomic state of the peduncular sheath in Acanthurus? Is there a glandular structure of any kind, and does it carry venom-like transcripts? (Question 3 is the critical test of the venom hypothesis at the level of the caudal weapon.)
What is the mechanical performance envelope of the acanthurine scalpel — bending stiffness, edge sharpness, fracture toughness — and how does it correlate with observed use rate and body size across species?
Do repeated evolutionary transitions to zooplanktivory reduce the tempo of weapon-elaboration, and if so, does the same transition also reduce the tempo of peduncular-contrast evolution — or does the coloration module continue to evolve for reasons unrelated to weapon use?

Three especially strong, publishable comparative-study designs.

Design A — Weapon geometry × territoriality × body-form phylogenetic analysis. Museum-based digitisation of scalpel outline and micro-CT scan for the entire family (target: 60+ species), yielding scalpel length, tip radius, edge curvature, and second moment of area. Combined with published territoriality scores [@robertson1986; @craig1996; @choat1985], body-shape data [@lungstrom2025; @friedman2016], and the Sorenson et al. [@sorenson2013] tree, run phylogenetic path analysis to test H1 vs H3 vs H4 vs H8. This is the highest-leverage single design because it uses existing museum material and existing molecular phylogenies, adds a new morphometric dataset, and directly tests the four most-cited hypotheses.

Design B — Localised peduncular-contrast phylogenetic analysis with receiver visual modelling. Reflectance spectrometry across >30 species, with visual modelling under conspecific + serranid + labrid pigment complements. Extract JND-based contrasts (socket vs body). Test whether localised contrast covaries with territoriality, weapon size, and habitat openness, controlling for whole-body colour diversity. This is the direct pre-registered test of the “localised vs whole-body contrast” question.

Design C — Definitive venom audit. Fresh-tissue histology and RNA-seq of the peduncular sheath in A. leucosternon, A. lineatus, A. nigrofuscus, P. laticlavius, and N. unicornis, with a small biochemical assay of extract cytotoxicity/haemolysis. This closes the venom question one way or the other, and produces high-visibility publications either as a null result (well-designed exclusion) or as a positive discovery.

Candid assessment: syndrome or modules?

The pattern that emerges from the primary literature does not support an integrated “colour–scalpel–venom” syndrome. The venom leg of the triad has no primary-literature support: there is no known acanthurid venom apparatus. The colour and scalpel legs are correlated in a subset of territorial Acanthurus, but the correlation is neither universal (many acanthurids have conspicuous coloration without heavy territorial weapon use, and vice versa) nor causally established. Repeated planktivory transitions [@friedman2016] evidently decouple the weapon-relevant selection regime from the coloration-relevant selection regime, and Prionurus (multiple keeled plates without conspicuous peduncular coloration) and Naso (fixed plates with highly variable peduncular coloration and pronounced sexual dimorphism only in body horn/rostrum but also plate width) each break the syndrome in a different direction.

The comparative evidence favours partially independent evolutionary modules: a weapon module (mobility, number, edge geometry, deployment angle) that tracks territoriality and interference-competition; a coloration module (localised peduncular contrast) that tracks receiver-specific signalling contexts and possibly aposematism; and, apparently, an absent-venom module — a “negative” trait that Acanthuridae share with their closest relatives except Siganidae. A rigorous phylogenetic comparative program will describe how loosely those modules covary, and where the exceptions (Prionurus, Zebrasoma scopas morphotypes, Naso unicornis dimorphism) sit relative to the mainstream Acanthurinae pattern. Framing the project as a test of modularity vs syndromic integration is, I think, the cleanest way to write the story up.

12 References

Bannikov, A. F., & Tyler, J. C. (1995). Phylogenetic revision of the fish families Luvaridae and †Kushlukiidae (Acanthuroidei), with a new genus and two new species of Eocene luvarids. Smithsonian Contributions to Paleobiology, 81, 1–45. https://doi.org/10.5479/si.00810266.81.1

Bellwood, D. R., & Tebbett, S. B. (2018). Unusual caudal spines in the surgeonfish Zebrasoma scopas. Coral Reefs, 37(1), 251. https://doi.org/10.1007/s00338-017-1652-z

Bellwood, D. R., Tebbett, S. B., & Bellwood, O. (2018). The role of the reef flat in coral reef trophodynamics: Past, present, and future. Ecology and Evolution, 8(8), 4108–4119. https://doi.org/10.1002/ece3.3967

Bernal, M. A., & Rocha, L. A. (2011). Acanthurus tractus Poey, 1860, a valid western Atlantic species of surgeonfish (Teleostei, Acanthuridae), distinct from Acanthurus bahianus Castelnau, 1855. Zootaxa, 2905(1), 60–68. https://doi.org/10.11646/zootaxa.2905.1.5

Bernardi, G., Nelson, P. A., Paddack, M. J., Rocha, L. A., & Craig, M. T. (2018). Genomic islands of divergence in the Yellow Tang and the Brushtail Tang surgeonfishes. Ecology and Evolution, 8(17), 8676–8685. https://doi.org/10.1002/ece3.4417

Carnevale, G., & Tyler, J. C. (2024). A new moorish idol (Teleostei, Zanclidae) from the Eocene of Bolca, Italy. Rivista Italiana di Paleontologia e Stratigrafia, 130(1). https://doi.org/10.54103/2039-4942/21794

Catano, L. B., Shantz, A. A., & Burkepile, D. E. (2014). Predation risk, competition, and territorial damselfishes as drivers of herbivore foraging on Caribbean coral reefs. Marine Ecology Progress Series, 511, 193–207. https://doi.org/10.3354/meps10921

Catano, L. B., Rojas, M. C. V., Malossi, R. J., et al. (2015). Reefscapes of fear: predation risk and reef heterogeneity interact to shape herbivore foraging behaviour. Journal of Animal Ecology, 85(1), 146–156. https://doi.org/10.1111/1365-2656.12440

Choat, J. H., & Bellwood, D. R. (1985). Interactions amongst herbivorous fishes on a coral reef: influence of spatial variation. Marine Biology, 89(3), 221–234. https://doi.org/10.1007/BF00393655

Craig, P. C. (1996). Intertidal territoriality and time-budget of the surgeonfish, Acanthurus lineatus, in American Samoa. Environmental Biology of Fishes, 46(1), 27–36. https://doi.org/10.1007/BF00001695

DeMartini, E. E. (2016). Sexual dimorphisms in the bluespine unicornfish, Naso unicornis (Acanthuridae): external metrics for movement ecology and life history. Copeia, 104(2), 498–505. https://doi.org/10.1643/CE-15-270

Emlen, D. J. (2008). The evolution of animal weapons. Annual Review of Ecology, Evolution, and Systematics, 39, 387–413. https://doi.org/10.1146/annurev.ecolsys.39.110707.173502

Ford, A. K., Bejarano, S., Marshell, A., & Mumby, P. J. (2016). Linking the biology and ecology of key herbivorous unicornfish to fisheries management in the Pacific. Aquatic Conservation: Marine and Freshwater Ecosystems, 26(4), 790–805. https://doi.org/10.1002/aqc.2623

Friedman, S. T., Price, S. A., Hoey, A. S., & Wainwright, P. C. (2016). Ecomorphological convergence in planktivorous surgeonfishes. Journal of Evolutionary Biology, 29(5), 965–978. https://doi.org/10.1111/jeb.12837

Goatley, C. H. R., Wroe, S., Tebbett, S. B., & Bellwood, D. R. (2018). An evaluation of a double-tailed deformity in a coral-reef surgeonfish Acanthurus nigrofuscus (Acanthuridae) using micro-computed tomography. Journal of Fish Biology, 92(5), 1645–1650. https://doi.org/10.1111/jfb.13600

Horodysky, A. Z., Brill, R. W., Crawford, K. C., Seagroves, E. S., & Johnson, A. K. (2013). Comparative visual ecophysiology of mid-Atlantic temperate reef fishes. Biology Open, 2(12), 1371–1381. https://doi.org/10.1242/bio.20136825

Klanten, O. S., van Herwerden, L., Choat, J. H., & Blair, D. (2004). Patterns of lineage diversification in the genus Naso (Acanthuridae). Molecular Phylogenetics and Evolution, 32(1), 221–235. https://doi.org/10.1016/j.ympev.2003.11.008

Konow, N., Bellwood, D. R., Wainwright, P. C., & Kerr, A. M. (2008). Evolution of novel jaw joints promote trophic diversity in coral reef fishes. Biological Journal of the Linnean Society, 93(3), 545–555. https://doi.org/10.1111/j.1095-8312.2007.00893.x

Krone, R., Bshary, R., Paster, M., Eisinger, M., van Treeck, P., & Schuhmacher, H. (2008). Defecation behaviour of the Lined Bristletooth Surgeonfish Ctenochaetus striatus (Acanthuridae). Coral Reefs, 27(3), 619–622. https://doi.org/10.1007/s00338-008-0365-8

Ludt, W. B., Rocha, L. A., Chakrabarty, P., & Bernal, M. A. (2019). The first complete mitochondrial genomes of sawtail surgeonfishes (Acanthuridae: Prionurus). Mitochondrial DNA Part B, 5(1), 212–213. https://doi.org/10.1080/23802359.2019.1699465

Lungstrom, L. L., Farjo, M., Isdonas, R., et al. (2025). Phylogenetics, trait covariance analysis, and the evolution of fin and body shape in the surgeonfishes. bioRxiv. https://doi.org/10.1101/2025.10.09.680739 (preprint; not peer-reviewed)

Marshall, N. J., Stieb, S. M., Cortesi, F., et al. (2023). Long-wavelength-sensitive (lws) opsin gene expression, foraging and visual communication in coral reef fishes. Molecular Ecology, 32(7), 1656–1672. https://doi.org/10.1111/mec.16831

Miyake, S., Ngugi, D. K., & Stingl, U. (2015). Diet strongly influences the gut microbiota of surgeonfishes. Molecular Ecology, 24(3), 656–672. https://doi.org/10.1111/mec.13050

Myrberg, A. A., Montgomery, W. L., & Fishelson, L. (1988). The reproductive behavior of Acanthurus nigrofuscus (Forskål) and other surgeonfishes (Fam. Acanthuridae) off Eilat, Israel (Gulf of Aqaba, Red Sea). Ethology, 79(1), 31–61. https://doi.org/10.1111/j.1439-0310.1988.tb00698.x

Ndandala, C. B., Mustapha, U. F., Wang, Y., et al. (2023). The perspective of fish venom: an overview of the physiology, evolution, molecular and genetics. Frontiers in Marine Science, 10, 1085669. https://doi.org/10.3389/fmars.2023.1085669

Palaoro, A. V., & Peixoto, P. E. C. (2022). The hidden links between animal weapons, fighting style, and their effect on contest success: a meta-analysis. Biological Reviews, 97(5), 1948–1966. https://doi.org/10.1111/brv.12877

Robertson, D. R., & Gaines, S. D. (1986). Interference competition structures habitat use in a local assemblage of coral reef surgeonfishes. Ecology, 67(5), 1372–1383. https://doi.org/10.2307/1938693

Schober, U. M., & Ditrich, H. (1992). Anatomy and use of the caudal spines in the aggressive behaviour of a surgeonfish (Osteichthyes: Acanthuridae). Marine Behaviour and Physiology, 21(4), 277–284. https://doi.org/10.1080/10236249209378831

Sivan, G. (2009). Fish venom: pharmacological features and biological significance. Fish and Fisheries, 10(2), 159–172. https://doi.org/10.1111/j.1467-2979.2008.00309.x

Smith, W. L., & Wheeler, W. C. (2006). Venom evolution widespread in fishes: a phylogenetic road map for the bioprospecting of piscine venoms. Journal of Heredity, 97(3), 206–217. https://doi.org/10.1093/jhered/esj034

Sorenson, L., Santini, F., Carnevale, G., & Alfaro, M. E. (2013). A multi-locus timetree of surgeonfishes (Acanthuridae, Percomorpha), with revised family taxonomy. Molecular Phylogenetics and Evolution, 68(1), 150–160. https://doi.org/10.1016/j.ympev.2013.03.014

Stankowich, T. (2012). Armed and dangerous: predicting the presence and function of defensive weaponry in mammals. Adaptive Behavior, 20(1), 32–43. https://doi.org/10.1177/1059712311426798

Stankowich, T., & Campbell, L. A. (2016). Living in the danger zone: exposure to predators and the evolution of spines and body armor in mammals. Evolution, 70(7), 1501–1511. https://doi.org/10.1111/evo.12961

Stankowich, T., Haverkamp, P. J., & Caro, T. (2014). Ecological drivers of antipredator defenses in carnivores. Evolution, 68(5), 1415–1425. https://doi.org/10.1111/evo.12356

Taylor, B. M., Rhodes, K. L., Marshell, A., & McIlwain, J. L. (2014). Age-based demographic and reproductive assessment of orangespine Naso lituratus and bluespine Naso unicornis unicornfishes. Journal of Fish Biology, 85(3), 901–916. https://doi.org/10.1111/jfb.12479

Tebbett, S. B., Goatley, C. H. R., & Bellwood, D. R. (2017). Clarifying functional roles: algal removal by the surgeonfishes Ctenochaetus striatus and Acanthurus nigrofuscus. Coral Reefs, 36(3), 803–813. https://doi.org/10.1007/s00338-017-1571-z

Tripalo, K., Japundžić, S., Sremac, J., et al. (2016). First record of Acanthuridae (surgeonfish) from the Miocene deposits of the Medvednica Mt. Geologia Croatica, 69(2), 201–204. https://doi.org/10.4154/gc.2016.15

Tyler, J. C., & Micklich, N. (2011). A new genus and species of surgeon fish (Perciformes, Acanthuridae) from the Oligocene of Kanton Glarus, Switzerland. Swiss Journal of Palaeontology, 130(2), 203–216. https://doi.org/10.1007/s13358-011-0016-5

Winterbottom, R. (1971). Movement of the caudal spine of some surgeonfishes (Acanthuridae, Perciformes). Copeia, 1971(3), 562. https://doi.org/10.2307/1442461

Winterbottom, R. (1993). Myological evidence for the phylogeny of recent genera of surgeonfishes (Percomorpha, Acanthuridae), with comments on the Acanthuroidei. Copeia, 1993(1), 21–39. https://doi.org/10.2307/1446292

Winterbottom, R., & McLennan, D. A. (1993). Cladogram versatility: evolution and biogeography of acanthuroid fishes. Evolution, 47(5), 1557–1571. https://doi.org/10.1111/j.1558-5646.1993.tb02175.x

Winterbottom, R., Tang, K. L., & Berendzen, P. B. (1999). The phylogenetic relationships of the suborder Acanthuroidei (Teleostei: Perciformes) based on molecular and morphological evidence. Molecular Phylogenetics and Evolution, 11(3), 415–425. https://doi.org/10.1006/mpev.1998.0596

Zheng, J., He, Y., Li, S., et al. (2026). First record of a Naso tergus specimen exhibiting unilateral absence of the shield-shaped bony plates on the caudal peduncle. Research Square (preprint). https://doi.org/10.21203/rs.3.rs-9537167/v1 (preprint; not peer-reviewed)

12.1 Un-verified sources referenced in the review (flagged)

The following works are cited or discussed in the narrative text or in secondary reviews I relied on, but I could not independently verify all bibliographic fields against a primary DOI/publisher record in the time available. They are listed here for transparency; readers relying on them should confirm directly.

Randall, J. E. (2001). Surgeonfishes of the world. Bishop Museum Press & Mutual Publishing, Honolulu. — This monograph is the standard taxonomic and natural-history reference for the family. I could not access its full text within this review’s window. Where I refer to it in the narrative it is as a source that is universally cited but that I have not independently checked page-by-page. (This is not a DOI-bearing work.)
Emlen, D. J. (2014). Animal Weapons: The Evolution of Battle. Henry Holt & Co. — Trade/academic book. Not a DOI-bearing work. The peer-reviewed Emlen 2008 Annual Review [@emlen2008] captures the same intellectual framework and has been verified.
Halstead, B. W. (1988). Poisonous and Venomous Marine Animals of the World (rev. ed.). Darwin Press. — Standard reference in medical toxicology, not a DOI-bearing work. Extensively cited via secondary reviews [@smith2006; @sivan2009; @ndandala2023].
Guiasu, R. C. & Winterbottom, R. (1993). Osteological evidence for the phylogeny of recent genera of surgeonfishes (Percomorpha, Acanthuridae). — Referenced through Bannikov & Tyler [@bannikov1995]; primary DOI not located in this review.

13 Reference-verification appendix

The table below lists every source I cite in the text, the DOI or persistent identifier used to verify it, the authoritative source that returned that verification, and any unresolved concerns.

Citation key	Verified DOI / stable ID	Authoritative verification source	Unresolved issues
Bannikov & Tyler 1995	10.5479/si.00810266.81.1	Smithsonian Contributions to Paleobiology; scite metadata	None
Bellwood & Tebbett 2018 (Z. scopas)	10.1007/s00338-017-1652-z	Coral Reefs; scite metadata	None
Bellwood, Tebbett & Bellwood 2018	10.1002/ece3.3967	Ecology and Evolution; scite metadata	None
Bernal & Rocha 2011	10.11646/zootaxa.2905.1.5	Zootaxa; scite metadata	None
Bernardi et al. 2018	10.1002/ece3.4417	Ecology and Evolution; scite metadata	None
Carnevale & Tyler 2024	10.54103/2039-4942/21794	Rivista Italiana di Paleontologia e Stratigrafia; scite metadata	None
Catano et al. 2014	10.3354/meps10921	Marine Ecology Progress Series; scite metadata	None
Catano et al. 2015	10.1111/1365-2656.12440	Journal of Animal Ecology; scite metadata	None
Choat & Bellwood 1985	10.1007/BF00393655	Marine Biology; scite metadata	None
Craig 1996	10.1007/BF00001695	Springer record; scite metadata	scite metadata mislabels the journal as “Journal of Applied Phycology”; the paper is Environmental Biology of Fishes 46(1). Journal fixed in the reference list; DOI is correct.
DeMartini 2016	10.1643/CE-15-270	Copeia; scite metadata; open PDF at NOAA IR	None
Emlen 2008	10.1146/annurev.ecolsys.39.110707.173502	Annual Reviews; scite metadata	None
Ford et al. 2016	10.1002/aqc.2623	Wiley; scite metadata	None
Friedman et al. 2016	10.1111/jeb.12837	Journal of Evolutionary Biology; scite metadata	None
Goatley et al. 2018	10.1111/jfb.13600	Journal of Fish Biology; scite metadata	None
Horodysky et al. 2013	10.1242/bio.20136825	Biology Open; scite metadata	None
Klanten et al. 2004	10.1016/j.ympev.2003.11.008	MPE; scite metadata	None
Konow et al. 2008	10.1111/j.1095-8312.2007.00893.x	BJLS; scite metadata	None
Krone et al. 2008	10.1007/s00338-008-0365-8	Coral Reefs; scite metadata	None
Ludt et al. 2019	10.1080/23802359.2019.1699465	Mitochondrial DNA Part B; scite metadata	None
Lungstrom et al. 2025	10.1101/2025.10.09.680739	bioRxiv; scite metadata	Preprint; peer review pending
Marshall et al. 2023	10.1111/mec.16831	Molecular Ecology; scite metadata	None
Miyake et al. 2015	10.1111/mec.13050	Molecular Ecology; scite metadata	None
Myrberg et al. 1988	10.1111/j.1439-0310.1988.tb00698.x	Ethology; scite metadata	None
Ndandala et al. 2023	10.3389/fmars.2023.1085669	Frontiers in Marine Science; scite metadata	None
Palaoro & Peixoto 2022	10.1111/brv.12877	Biological Reviews; scite metadata	None
Robertson & Gaines 1986	10.2307/1938693	Ecology; scite metadata	Title in scite metadata contains a typo (“Compoetition”); confirmed original title uses “Competition”.
Schober & Ditrich 1992	10.1080/10236249209378831	Taylor & Francis DOI record; multiple secondary citations quoting content	Full text not directly accessed in this review; behavioural claim quoted through [@bellwood2017scopas] and [@zheng2026]
Sivan 2009	10.1111/j.1467-2979.2008.00309.x	Fish and Fisheries; scite metadata	None
Smith & Wheeler 2006	10.1093/jhered/esj034	Journal of Heredity; scite metadata; open PDF at AMNH	scite metadata lists author order as “Wheeler, Smith”; publisher record shows “Smith & Wheeler”. Author order fixed in bibliography.
Sorenson et al. 2013	10.1016/j.ympev.2013.03.014	MPE; scite metadata; PubMed 23542000	None
Stankowich 2012	10.1177/1059712311426798	Adaptive Behavior; scite metadata	None
Stankowich et al. 2014	10.1111/evo.12356	Evolution; scite metadata	None
Stankowich & Campbell 2016	10.1111/evo.12961	Evolution; scite metadata	None
Taylor et al. 2014	10.1111/jfb.12479	Journal of Fish Biology; scite metadata	None
Tebbett et al. 2017	10.1007/s00338-017-1571-z	Coral Reefs; scite metadata	None
Tripalo et al. 2016	10.4154/gc.2016.15	Geologia Croatica; scite metadata	None
Tyler & Micklich 2011	10.1007/s13358-011-0016-5	Swiss Journal of Palaeontology; scite metadata	None
Winterbottom 1971	10.2307/1442461	Copeia via JSTOR; scite metadata	None
Winterbottom 1993	10.2307/1446292	Copeia via JSTOR; scite metadata	None
Winterbottom & McLennan 1993	10.1111/j.1558-5646.1993.tb02175.x	Evolution / OUP article page directly fetched	None
Winterbottom et al. 1999	10.1006/mpev.1998.0596	MPE; scite metadata	None
Zheng et al. 2026	10.21203/rs.3.rs-9537167/v1	Research Square; scite metadata	Preprint; peer review pending

Un-verified sources. Randall 2001, Emlen 2014, Halstead 1988, and Guiasu & Winterbottom 1993 do not have primary DOIs verified in this review. They are widely cited monographs / books and their content is discussed only where independent primary sources confirm the substantive claim.