The muscles serving the ventral portion of the gill arches ( = infrabranchial musculature) are poorly known in bony fishes. A comparative analysis of the infrabranchial muscles in the major percomorph lineages reveals a large amount of phylogenetically-relevant information. Characters derived from this anatomical system are identified and discussed in light of current hypotheses of phylogenetic relationships among percomorphs. New evidence supports a sister-group relationship between the Batrachoidiformes and Lophiiformes and between the Callionymoidei and Gobiesocoidei. Investigated data also corroborate the existence of two monophyletic groups, one including the Pristolepididae, Badidae, and Nandidae, and a second clade consisting of all non-amarsipid stromateiforms. New synapomorphies are proposed for the Atherinomorphae, Blenniiformes, Lophiiformes, Scombroidei (including Sphyraenidae), and Gobiiformes. Within the latter order, the Rhyacichthyidae and Odontobutidae are supported as the successive sister families of all remaining gobiiforms. The present analysis further confirms the validity of infrabranchial musculature characters previously proposed to support the grouping of the Mugiliformes with the Atherinomorphae and the monophyly of the Labriformes with the possible inclusion of the Pholidichthyiformes. Interestingly, most hypotheses of relationships supported by the infrabranchial musculature have been advanced by preceding anatomists on the basis of distinct data sources, but were never recovered in recent molecular phylogenies. These conflicts clearly indicate the current unsatisfactory resolution of the higher-level phylogeny of percomorphs.
Citation: Datovo A, de Pinna MCC, Johnson GD (2014) The Infrabranchial Musculature and Its Bearing on the Phylogeny of Percomorph Fishes (Osteichthyes: Teleostei). PLoS ONE 9(10): e110129. https://doi.org/10.1371/journal.pone.0110129
Editor: Andreas Hejnol, Sars International Centre for Marine Molecular Biology, Norway
Received: September 10, 2013; Accepted: September 17, 2014; Published: October 13, 2014
Copyright: © 2014 Datovo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was funded by a Postdoctoral Fellowship from São Paulo Research Foundation (FAPESP; grant #2010/18984-9). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The division Percomorphacea (sensu Wiley and Johnson ) encompasses more than half of all species of living teleostean fishes and one quarter of extant vertebrates (Fig. 1; , . In the recent classification of Wiley and Johnson , based on morphological evidence, ca. 17,000 extant species of percomorphs are distributed into 30 orders, 23 of which are placed in a vast unresolved basal polytomy (Fig. 1). Resolution of the interrelationships among the major percomorph groups remains one of the most daunting challenges in the systematics of the Teleostei. While several recent molecular studies pursued the problem –, few comparative anatomical studies across a broad range of the Percomorphacea have been performed in the last decade .
While understanding of the intrarelationships of percomorphs is undoubtedly deficient, it is also true that several morphological complexes in the group, especially those from soft anatomy, remain severely under-studied. Among the morphological synapomorphies for the major percomorph groups listed in Wiley and Johnson , only 5% are from the skeletal musculature (Fig. 2). Despite important contributions to the knowledge of percomorph myology , definition of the major groups within this clade is still massively based on osteological characters (Fig. 2).
Synapomorphies and classification based on Wiley and Johnson .
The branchial skeleton of fishes has been extensively studied by systematists especially after a series of studies by G. Nelson on that system –. More recently, Springer and Johnson  published a seminal monograph focusing on the dorsal and some posteroventral branchial muscles of bony fishes. In contrast, most muscles serving the ventral portion of the branchial arches have received scant attention from systematists (: p. 294). To facilitate communication, the muscles associated with the ventral skeletal elements of the gill arches (basibranchials, hypobranchials, and ceratobranchials) are hereafter denominated infrabranchial muscles (from Latin infra meaning below, under). Data on the infrabranchial musculature of teleosts is sparse in the literature, with most studies on the subject dealing with only a single or a few closely related species –.
This study presents a comparative analysis of the infrabranchial musculature of 90 representatives of all 30 percomorph orders. Our aim is to provide a first overview of the main patterns of variation in the infrabranchial musculature of percomorphs, a subject so far nearly unexplored. Although the taxonomic density of our sampling can and should eventually be increased, it is broad enough to provide a robust comparative framework that demonstrates the existence of significant phylogenetically-correlated information previously unreported, representing a number of new characters bearing on the higher-level phylogeny of the Percomorphacea. We expect our results can lay the foundations for further explorations into the infrabranchial myology of the group.
Materials and Methods
Higher-level classification of the Teleostei follows Wiley and Johnson . Families not listed in that study are as recognized in Eschmeyer and Fong . Current taxonomic validity of the cited species follows Eschmeyer .
The research employed only ethanol-preserved specimens deposited in museums and did not involve animal experimentation or examination of fossil specimens. No permits were required for the described study, which complied with all relevant regulations. Preserved specimens were double-stained for cartilage and bone prior to dissection following the protocol of Datovo and Bockmann . Examined material is listed in Table S1 and is deposited in the following institutions: Humboldt State University Fish Collection, USA (HSU); Laboratório de Ictiologia de Ribeirão Preto, Universidade de São Paulo, Brazil (LIRP); Museum of Comparative Zoology, Harvard University, USA (MCZ); Museu de Zoologia da Universidade de São Paulo, Brazil (MZUSP); Scripps Institution of Oceanography, USA (SIO); and National Museum of Natural History, Smithsonian Institution, USA (USNM). Access to material of those collections was authorized by respective curators. Specimens were examined at their original institutions or loaned to MZUSP or USNM between 01 May 2011 and 03 March 2014.
The term insertion refers to the attachment of the muscle to the structure (usually a bone) that presumably moves (or moves more intensely) during its contraction; origin is defined as the opposite muscle attachment to the stationary (or less movable) skeletal element . Musculous attachment (origin or insertion) is when the muscles fibers attach directly to the skeleton without the mediation of any macroscopically evident tendon. In the tendinous attachment, the muscle fibers converge onto a macroscopically evident tendon, which, in turn, attaches to the skeleton. In some instances, the attachment of a muscle is partially musculous and partially tendinous.
Photographs were taken with a Leica DFC420 digital camera device attached to a Leica MZ16 stereomicroscope. Final images are composite multifocal montages derived from the stacking of several individual photographs taken at different focal planes. The multifocal montage was prepared in CombineZP . Montages were then retouched digitally in Adobe Photoshop CS6 in order to remove artefacts (bubbles, specks, etc.) and chromatic imperfections in the original anatomical preparations (e.g., muscles unduly stained with Alizarin red or Alcian blue) and to enhance structures of interest via adjustments in contrast, shadows and channel levels.
Results and Discussion
Generalized acanthopterygian pattern of the infrabranchial musculature
This section provides an overview of the general pattern of the infrabranchial musculature present in the majority of percomorph lineages. This pattern is also shared by examined representatives of other acanthopterygian orders (i.e., Stephanoberyciformes, Zeiformes, and Beryciformes) and is thus interpreted as being plesiomorphic for the Percomorphacea. The generalized acanthopterygian pattern is presented as a baseline for the description and discussion of variations specific to smaller subgroups.
Obliqui ventrales I, II, and III are bilaterally-paired muscles sequentially arranged from first to third branchial arches, respectively (Fig. 3). Each obliquus ventralis has a broad origin on the ventral surface of the hypobranchial and a smaller area of insertion on the anteroventral region of the ceratobranchial of the same arch.
MZUSP 67283. Ventral view; left pharyngoclaviculares externus and internus, and right rectus communis removed.
Rectus ventralis IV originates from the ventral face of hypobranchial 3 and inserts on the anteroventral portion of ceratobranchial 4 (Fig. 3).
The muscle fibers of obliquus ventralis III and rectus ventralis IV may attach directly to hypobranchial 3 or originate from this bone via a semicircular ligament (Fig. 4). When present, this ligament encircles posteriorly the anterior portion of the ventral aorta, which splits into left and right branches after its passage through the semicircular ligament. The presence of the semicircular ligament is likely plesiomorphic for percomorphs , . However, both the presence of this ligament and its association with rectus ventralis IV and obliquus ventralis III exhibit a mosaic distribution across the Euteleosteomorpha (pers. obs., ), so that the precise phylogenetic interpretation of these features remains uncertain. In some instances, the sagittal portion of the semicircular ligament is attached to the basibranchials and this condition may confound the unequivocal recognition of that ligament.
MZUSP 71680. Ventral view; rectus communis removed.
Transversi ventrales IV and V are unpaired muscles that interconnect the anteroventral regions of the contralateral ceratobranchials 4 and 5, respectively (Fig. 3). Either or both muscles may exhibit a sagittal raphe. In all examined lower acanthopterygians and most percomorphs, rectus ventralis V has a sagittal raphe and, consequently, this condition is hypothesized to be plesiomorphic for percomorphs. Non-acanthopterygian percomorphs either have or lack a sagittal raphe in rectus ventralis IV, so that the polarity of this character is uncertain.
The rectus communis originates from the urohyal and inserts via a tendon on the lateral aspects of ceratobranchial 5 (Fig. 3). The anterior portion of the muscle is partially continuous with the sternohyoideus, which runs from the urohyal to the cleithrum. Lauder  proposed that the total or partial association of the rectus communis with the urohyal is a synapomorphy for the Ctenosquamata, and Stiassny  reported that this muscle has completely lost its association with hypobranchial 3 in holacanthopterygians (including percomorphs). However, a rectus communis with partial or total origin on hypobranchial 3 is present in most non-percomorph acanthopterygians and a few percomorphs herein examined. As a consequence, it is unclear whether or not the attachment of the rectus communis to hypobranchial 3 is plesiomorphic for percomorphs (see also “Modifications of unclear systematic value” below).
The pharyngoclavicularis externus originates from the anteroventral region of the cleithrum and inserts onto the anterolateral aspects of the ventral face of ceratobranchial 5 (Fig. 3). Both the origin and insertion of the muscle are primitively musculous.
The pharyngoclavicularis internus arises from the posterior portion of the anterodorsal face of the cleithrum and inserts on the anteroventral region of ceratobranchial 5 (Fig. 3). Near its insertion, the muscle passes medial to the pharyngoclavicularis externus.
The sternohyoideus originates from the anteroventral portion of the cleithrum, coracoid, and the myoseptum shared with the anteriormost segment of the obliquus inferioris ( = ventral division of the hypaxialis). The sternohyoideus is often segmented into three myomeres and inserts anteriorly onto the urohyal. A vertical membranous wall of connective tissue arises from the sagittal plane of the sternohyoideus and attaches dorsally to the ventral surface of the gill arches, thus dividing the branchial chamber beneath the arches into left and right halves. A strengthened oblique band of collagen embedded into the body of this membrane is often present between hypobranchial 3 and the dorsomedial portion of the sternohyoideus. The degree of differentiation of this strengthened band greatly varies across the examined acanthopterygians and the well-differentiated band of some taxa serves as an additional insertional tendon connecting the sternohyoideus to hypobranchial 3.
The sphincter oesophagi is an unpaired muscle that encircles the anterior portion of the esophagus and attaches anteriorly to ceratobranchial 5 and epibranchial 4 (Fig. 3).
Among percomorphs, the basic infrabranchial muscular configuration described above is present at least in the generalized members of the Cottiformes , , Gasterosteiformes , Nototheniiformes, Ophidiiformes , Scorpaenoidei , , Serranoidei , and several perciform families – Apogonidae, Centrarchidae , Haemulidae, Lutjanidae, Percidae , , Sciaenidae , , and Sparidae .
Several modifications in the generalized acanthopterygian pattern of the infrabranchial musculature arose in different percomorph lineages. These changes are described and discussed below in light of current hypotheses of phylogenetic relationships of percomorphs. Discussions of putative synapomorphies for the investigated groups are preceded by sequential numbers in bold type. Derived conditions that are apparently autapomorphic for single examined species or other lower taxonomic levels are outside the scope of this paper and are not included.
Badidae, Nandidae, and Pristolepididae
Several morphologists have long suggested a hypothesis of close relationship between the Anabantoidei, Channoidei, Pristolepididae, Badidae, and Nandidae (Fig. 5A) , , –. Springer and Johnson  coined the term Anabantomorpha to refer to the group and its monophyly was recently corroborated in two recent molecular analyses , . While the Anabantoidei and Channoidei are often considered sister taxa ( = Anabantiformes sensu Wiley and Johnson ; = Labyrinthici Regan ), the relationships among the Pristolepididae, Badidae, and Nandidae vary among different studies. These three families were grouped into a monophyletic lineage in Near et al. , an arrangement corroborated by the following specialization of the infrabranchial musculature (Fig. 5A).
Characters superimposed on cladograms of the relationships among the (A) Anabantomorpha , ; (B) Batrachoidiformes and Lophiiformes , , ; (C) Callionymoidei and Gobiesocoidei , –; (D) Gobiiformes –; (E) Mugiliformes and Atherinomorphae , ; (F) Scombroidei, Pomatomidae and Scombrolabracidae ; and (G) Stromateiformes . Characters numbered as in the text.
- 1. In the plesiomorphic condition for the Percomorphacea, rectus ventralis IV inserts solely on ceratobranchial 4 (Figs. 3, 4). In Pristolepis, Badis, and Nandus this muscle has an additional insertion on the anterior region of ceratobranchial 5 (Fig. 6). This derived condition is unique among the examined acanthopterygians and thus supports a pristolepidid-badid-nandid alignment (Fig. 5A).
Batrachoidiformes and Lophiiformes
Several previous morphological studies proposed that the Batrachoidiformes and Lophiiformes form a monophyletic group, often termed Pediculati (Fig. 5B) –. This hypothesis has been repeatedly contested by molecular phylogenies that allocate these two orders in clades far removed from each other by several intercalated acanthopterygian taxa , , , . The present study identified some unique derived myological characters shared by batrachoidiforms and lophiiforms.
- 2. Acanthopterygians primitively have a well-developed obliquus ventralis I muscle (Figs. 3, 4, 6). All lophiiforms and batrachoidiforms except Chaunax and Porichthys exhibit an extreme reduction of this muscle. Obliquus ventralis I is an almost imperceptible vestigial muscle in Batrachoides and Antennarius. The batrachoidiforms Thalassophryne and Triathalassothia and the lophiiforms Lophius and Ogcocephalus completely lack obliquus ventralis I.
- 3. Primitively in percomorphs, the rectus communis inserts solely on ceratobranchial 5 (Fig. 3). All examined batrachoidiforms and lophiiforms have a rectus communis with a medial section that retains the plesiomorphic insertion on ceratobranchial 5 and an additional lateral section that attaches to ceratobranchial 4 (Fig. 7). These sections are continuous with each other at their origins in batrachoidiforms and the lophiiforms Antennarius and Chaunax, whereas in Lophius and Ogcocephalus the two sections are completely separated along their entire extent. In batrachoidiforms the insertion of the rectus communis on ceratobranchial 4 is achieved via a tendon that partially merges dorsally with the insertional tendon of rectus ventralis IV (Fig. 7).
MZUSP 46971. Ventral view; right pharyngoclavicularis externus, pharyngoclavicularis internus and rectus communis removed.
The two derived characters above clearly favor the hypothesis previously advanced by some morphologists that the Batrachoidiformes and Lophiiformes are sister taxa (Fig. 5B) –. Under that scheme, these characters are optimized as synapomorphies for the Pediculati, with reversals of character 2 in Chaunax and Porichthys (Fig. 5B) – it is worth mentioning that neither porichthyines nor chaunacids have been proposed as primitive members of their respective orders , , , , . Most, but not all, batrachoidiforms and lophiiforms additionally share other specializations of the gill-arch system that are not found elsewhere among acanthopterygians, such as the extreme reduction of obliquus ventralis II (character 17), the musculous insertion of the rectus communis on ceratobranchial 5 (character 26), and the anterior tip of ceratobranchial 1 positioned posterior to that of ceratobranchial 2 (Fig. 7; Pietsch and Orr : fig. 12B) – vs. anterior tip of ceratobranchial 1 anterior to that of ceratobranchial 2 (Fig. 3). However, the precise optimizations of such characters are problematic and demand a more encompassing taxonomic sampling of the Pediculati. On the molecular phylogenies of the Acanthopterygii , , , all foregoing similarities would be mapped as convergent in batrachoidiforms and lophiiforms. These studies grouped lophiiforms with tetraodontiforms and caproiforms, rather than with batrachoidiforms. Although some phenotypic traits were recently offered as possibly corroborating the alignment of lophiiforms with tetraodontiforms , , such a hypothesis definitely gains no support from the myological data herein investigated.
Monophyly of the Lophiiformes is supported to date by six synapomorphies , . The following derived condition of the infrabranchial musculature is herein interpreted as an additional unique synapomorphy for the order (Fig. 5B).
- 4. The urohyal is the primitive site of origin for the rectus communis in ctenosquamates , . Lophiiforms lack a urohyal  and have a rectus communis originating from the dorsal hypohyal. In Ogcocephalus, the origin of the rectus communis involves both the dorsal hypohyal and the anteromedial region of the anterior ceratohyal.
Examined blenniiforms share the following modification in the infrabranchial musculature.
- 5. Primitively in percomorphs, transversus ventralis IV is an unpaired muscle that connects ceratobranchials 4 of each side (Figs. 3, 4, 7). In Enneanectes and Scartella the anteroventral portion of transversus ventralis IV retains these primitive attachments, whereas the posterodorsal portion of the muscle attaches to the anterolateral margin of ceratobranchial 5 (Fig. 8). This condition is also present in other blenniiform families  and was considered a synapomorphy for the order by Springer and Orrel (: character 35, state 1). Wiley and Johnson  incorrectly listed the absence of transversus ventralis IV ( = character 35, state 2 of Springer and Orrel ) as a synapomorphy for the Blenniiformes. Among all other examined percomorphs, only exocoetoids and labriforms exhibit a condition similar, but not identical, to that in blenniiforms. In exocoetoids and labriforms the dorsal portion or the entire transversus ventralis IV inserts on a conspicuous ventromedial keel of ceratobranchial 5, rather than on the anterolateral margin of ceratobranchial 5 (Fig. 9; see character 10 below).
MZUSP 60573. Ventral view; right pharyngoclavicularis externus and rectus communis removed.
Callionymoidei and Gobiesocoidei
Some traditional classifications considered gobiesocoids as closely related to batrachoidiforms and lophiiforms , , . That hypothesis was later contested by other morphological studies that presented evidence supporting the grouping of Gobiesocoidei (Gobiesocidae) with Callionymoidei (Draconettidae+Callionymidae; Fig. 5C) , , –. Molecular phylogenies, on the other hand, recovered gobiesocoids, callionymoids, batrachoidiforms, and lophiiforms as not closely related to each other , , , , . The following evidence from the infrabranchial myology supports a gobiesocoid-callionymoid alignment.
- 6. Rectus ventralis I is a muscle apparently derived from an anterolateral subdivision and/or expansion of obliquus ventralis I that acquires a novel attachment on the hyoid arch, usually the dorsal hypohyal (Fig. 8) . In gobiesocoids and callionymoids, rectus ventralis I is a flat muscle anterolaterally oriented toward its origin and with a broad insertion solely on the anterolateral border of the ossified portion of ceratobranchial 1 (Fig. 10) . Rectus ventralis I is a conspicuous muscle present on both sides of the examined gobiesocoid and callionymid, but it is very small and present only on the right side of the examined draconettid. All other known percomorphs either lack rectus ventralis I (Figs. 3, 6, 7, 9) or have this muscle anteromedially oriented towards its origin (Fig. 8) and usually with a narrow insertion on hypobranchial 1 and/or the anterior cartilaginous tip of ceratobranchial 1. The derived condition of rectus ventralis I of gobiesocoids and callionymoids thus supports the hypothesis that these taxa are sister groups (Fig. 5C) , , –.
MZUSP 66707. Ventral view; gill rakers and right rectus communis removed.
The examined gobiesocoid, Gobiesox strumosus, possesses recti ventrales II and III, which likely represent anterolateral subdivisions of obliqui ventrales II and III, respectively. The draconettid Centrodraco oregonus also has rectus ventralis II. Dietz  reported the presence of rectus ventralis II in Callionymus lyra, but this muscle is absent in the callionymid herein examined (Fig. 10). Some, but not all, members of the Blenniiformes also have recti ventrales II and III (Fig. 8). Because several independent studies proposed that blenniiforms and gobiesociforms form a monophyletic group , , , , the precise optimization of the presence of recti ventrales II and III in these taxa is ambiguous. To further complicate the issue, other apparently non-closely related percomorphs also have recti ventrales II and/or III: some acanthuriforms , , tetraodontiforms , cottoids , , and gasterosteiforms .
- 7. Examined gobiiforms (including the putatively basal Rhyacichthyidae) possess obliquus ventralis III divided into an anterodorsal section that directly arises from ceratobranchial 3 and a posteroventral section that originates from the semicircular ligament (Fig. 11). Contrarily, most acanthomorphs, including taxa commonly suggested as possible gobiiform closest relatives , , , have undivided obliquus ventralis III (Figs. 3, 4, 6, 9, 10). Among other examined acanthopterygians, a similarly divided obliquus ventralis III was found only in the batrachoidiforms Batrachoides and Porichthys. In view of the long phylogenetic distance between batrachoidiforms on the one hand and gobiiforms on the other, parsimony considerations indicate that the divided obliquus ventralis III is convergent in the two orders (Fig. 5B, D).
MZUSP 66368. Ventral view; gill rakers, rectus communis, and left pharyngoclavicularis externus, pharyngoclavicularis internus, and sternobranchialis removed.
Prior anatomical studies placed the Rhyacichthyidae and Odontobutidae as successive sister groups of the clade formed by all remaining gobiiforms , . Molecular phylogenies contrastingly grouped these families into a monophyletic lineage that is, in turn, sister to remaining gobiiforms , . The following newly discovered characters corroborate the former hypothesis advanced by morphologists (Fig. 5D).
- 8. Percomorphs primitively have an undivided obliquus ventralis II that runs from ceratobranchial 2 to hypobranchial 2 (Figs. 3, 6, 7, 9, 10). In all gobiiforms except rhyacichthyids, obliquus ventralis II is completely subdivided into an anterolateral section, which retains the plesiomorphic muscle attachments, and a posteromedial section, which originates from basibranchial 2+3 ( = O'2 muscle of Dietz ; Fig. 11). Among all other known acanthopterygians, a similar but not identical condition is found only in the Atherinomorphae and Gasterosteiformes . These taxa have an obliquus ventralis II originating from both hypobranchial 2 and basibranchial 2+3, but the muscle is never subdivided into two separate sections as it is in gobiiforms (Fig. 12). Furthermore, gobiiforms are apparently not closely related to either atherinomorphs or gasterosteiforms , , , , –. Consequently, the presence of a separate medial section of obliquus ventralis II originating from basibranchial 2+3 may be interpreted as an evidence for the monophyly of the non-rhyacichthyid gobiiforms (Fig. 5D).
- 9. Rhyacichthyids exhibit the primitive acanthopterygian condition in which the sternohyoideus originates from the cleithrum and inserts solely on the urohyal. In all remaining gobiiforms, a dorsomedial segment of the sternohyoideus attaches to the anteroventral tip of hypobranchial 3 (Fig. 11) . This muscle segment is partially separated from the main bulk of the sternohyoideus in odontobutids and completely separated from the sternohyoideus in the remaining gobiiforms. Following Winterbottom , this muscle segment attached to hypobranchial 3 may be designated as sternobranchialis ( = ‘Ad’ of Dietz ). In Dormitator, the insertional tendon of the sternobranchialis further extends anteriorly to attach to both hypobranchial 2 and 3. In an ordered series of transformation, the attachment of the dorsomedial fibers of the sternohyoideus to hypobranchial 3 (that is, an at least partially differentiated sternobranchialis) may thus be taken as a synapomorphy for the non-rhyacichthyid gobiiforms (state 1), whereas a fully separated sternobranchialis is considered a synapomorphy for the clade formed by all gobiiforms except the Rhyacichthyidae and Odontobutidae (state 2; Fig. 5D). Among percomorphs, a sternobranchialis is also present in tetraodontiforms and the caproiform Antigonia (Fig. 13), taxa that are distantly allied to gobiiforms , , , , , , . Additionally, the sternobranchialis in gobiiforms is markedly different from that of tetraodontiforms and caproiforms. The gobiiform sternobranchialis has a parallel arrangement of fibers along almost its entire extent, a nearly horizontal orientation, antimeres distant from each other, and a single undivided insertion on hypobranchial 3 or hypobranchials 2 and 3 (Fig. 11). In contrast, the sternobranchialis of caproiforms and tetraodontiforms is characterized by a bulged origin and strongly flattened insertion, a nearly vertical orientation, antimeres much closer to one another, and an insertional aponeurosis splitting into two or more tendons that attach to several bones of the branchial and/or hyoid arches (Fig. 13) . Although such differences in themselves are not decisive refutation of homology, they provide further corroboration that the gobiiform sternobranchialis evolved independently (Fig. 5D) of that in caproiforms and tetraodontiforms (see character 29).
MZUSP 67017. Ventral view; rectus communis and left pharyngoclaviculares externus and internus removed.
Labriformes and Pholidichthyiformes
In contrast to the situation in most other percomorph groups, the branchial musculature of labriforms has been extensively studied , , , , –. The enigmatic Pholidichthys, sole genus of Pholidichthyiformes, has been indecisively aligned with different percomorph groups, including blenniiforms, gobiiforms, labriforms, and trachiniforms , , –. Stiassny and Jensen  highlighted that labriforms and pholidichthyiforms apparently share several similarities in the branchial skeleton, and the analysis of Springer and Orrel  placed Pholidichthys within the Labriformes. The present study confirms that pholidichthyiforms exhibit a putative labriform synapomorphy.
- 10. In the primitive condition for percomorphs, transversus ventralis IV attaches solely to the contralateral ceratobranchials 4, completely bypassing ceratobranchial 5 (Figs. 3, 4, 6, 7, 10–12). In labriforms and pholidichthyiforms, at least the dorsal portion of transversus ventralis IV attaches to a conspicuous ventromedial keel of ceratobranchial 5 (Fig. 9) , , , , –. The same condition is also present in exocoetoid beloniforms, a group well nested within the Atherinomorphae. On the grounds of parsimony, the transversus ventralis IV muscle attaching to the ventral keel of ceratobranchial 5 of exocoetids is interpreted as having independently evolved from that of labriforms  and pholidichthyiforms.
As noted above (character 5), blenniiforms have only the posterior portion of transversus ventralis IV attached to the anterolateral region of ceratobranchial 5 (Fig. 8). In our opinion, this condition is clearly distinct from that in both labriforms, pholidichthyiforms, and exocoetids in which the entire, or only the dorsal part, of the muscle attaches to a ventromedial keel of ceratobranchial 5 (Fig. 9). Nevertheless, Springer and Orrel  coded these different conditions under the same character state: transversus ventralis IV attached to ceratobranchial 5 (their character 35, state 1). Under that codification, the analysis of Springer and Orrel  did not recover the insertion of transversus ventralis IV on ceratobranchial 5 as a synapomorphy for their clade 5, which comprises the Labriformes plus Pholidichthyiformes.
Kaufman and Liem  and Stiassny and Jensen  listed a second myological synapomorphy for the Labriformes: the sphincter oesophagi lacking a dorsal subdivision. The referred dorsal subdivision of the sphincter oesophagi corresponds to the “sphincter oesophagi division” or SOD muscle of Springer and Johnson , which passes dorsal to the insertion of the retractor dorsalis. Although the sphincter oesophagi was included in the present analysis, it seems that the unequivocal identification of SOD in some taxa necessarily requires the study of other dorsal branchial muscles, especially the transversus dorsalis posterior. The latter muscle occupies a position similar to that of SOD and those two muscles are sometimes continuous with each other (: p. 13). As the dorsal branchial muscles were not investigated in the present study, we could not positively determine the presence or absence of the SOD in examined percomorphs and must leave this question unresolved.
The monophyly of the Labriformes, with the possible inclusion of Pholidichthys , is currently supported only by characters from the branchial arches , , , , and the lack of evidence from other anatomical systems is often cited as a major criticism of the supposed monophyly of the order , . All molecular analyses to date have recovered a polyphyletic Labriformes but, interestingly, Pholidichthys is often resolved as sister to the labriform family Cichlidae , , , . Resolution of these conflicts is beyond the scope of the present study.
Mugiliformes and Atherinomorphae
Stiassny  proposed mugiliforms and atherinomorphs as sister taxa (Fig. 5E) on the basis of a number of shared morphological specializations, including the following character of the infrabranchial musculature.
- 11. In the primitive percomorph condition, the pharyngoclavicularis externus is an undivided muscle that inserts on the anteroventral region of ceratobranchial 5 (Figs. 3, 6, 7, 8, 9, 10, 11, 13). In mugiliforms and non-exocoetoid atherinomorphs, this muscle is dorsally divided into an anterior and a posterior section, which insert on the anteroventral and posteroventral regions of ceratobranchial 5, respectively (Fig. 12). The present study confirms the distribution and putative validity of this character as a synapomorphy for the clade Mugiliformes+Atherinomorphae (Fig. 5E). Exocoetoid beloniforms have a single undivided pharyngoclavicularis externus inserting solely on the posteroventral region of ceratobranchial 5, thus resembling the primitive state for the Percomorphacea. Stiassny (: p. 6) interpreted the exocoetoid condition as the result of a loss of the anterior subdivision of the pharyngoclavicularis externus, rather than a reversal to the ancestral percomorph state. We have not disclosed new evidence either for or against such interpretation.
Stiassny  proposed a second character of the infrabranchial musculature as supporting a mugiliform-atherinomorph relationship: the “pharyngohyoideus muscle [ = rectus communis] reduced to a small fan-shaped muscle with an elongate tendon”. This character was later discarded by Stiassny  because of the extensive degree of variation detected among a broader range of percomorphs, which renders the delimitation of discrete character-states problematic. Such observations and conclusions are supported by the present study.
The monophyly of the Atherinomorphae is almost unanimously accepted by phylogenetic analyses based on both genetic and phenotypic data sets (Fig. 5E) –, , , , –. The following derived condition was identified in the infrabranchial musculature of the examined atherinomorphs.
- 12. Percomorphs primitively have an undivided obliquus ventralis III originating from the ventral face of hypobranchial 3 and inserting on the ventrolateral region of ceratobranchial 3 (Figs. 3, 6–7, 9, 10, 13). In four of the five examined atherinomorphs, obliquus ventralis III is expanded dorsomedially in such a way that the insertion of the muscle involves ceratobranchial 3 and the dorsal face of hypobranchial 3, whereas its origin extends over basibranchial 2+3 and the ventral face of hypobranchial 3 (Fig. 12). This condition was not found elsewhere among percomorphs, and is thus hypothesized to be synapomorphic for the Atherinomorphae (Fig. 5E), with a reversal in the cyprinodontiform Hypsolebias, which exhibits the primitive percomorph condition.
The Scombroidei has long been recognized as a natural group by morphologists , , –, though the exact composition of the suborder has varied over time. In the morphological analysis of Johnson , the Scombroidei includes the Sphyraenidae and excludes the monotypic Scombrolabracidae and Pomatomidae, but the two later families are recovered as the successive sister groups of the suborder (Fig. 5F). Molecular analyses, however, resolve the Scombroidei as polyphyletic , , , , , . The following unique apomorphic condition of the infrabranchial musculature supports the monophyly of the Scombroidei sensu Johnson .
- 13. In most percomorphs, including Scombrolabrax and Pomatomus, the pharyngoclavicularis externus lacks any intermediate aponeurosis ( = a laminar, flattened tendon; Figs. 3, 6, 7, 8, 9, 10, 11, 12, 13). Most scombroids contrastingly have an intermediate aponeurosis at the mid-length of this muscle (Fig. 14). This aponeurosis runs across the entire anteroposterior expanse of the pharyngoclavicularis externus in sphyraenids and the scombrid Scomberomorus (M. Nakae, pers. comm.), but only at the anterior portion of the same muscle in gempylids, xiphiids, and the scombrids Euthynnus, Scomber, and Thunnus (M. Nakae, pers. comm.). Istiophorids exhibit a further modified condition in which a broad aponeurosis is located at the dorsal end (insertion) of the pharyngoclavicularis externus. Trichiurids, in turn, are the only investigated scombriforms lacking an aponeurosis in that muscle (, pers. obs.). Among all other examined acanthopterygians, only Kyphosus also possesses an intermediate aponeurosis at the anterior region of the pharyngoclavicularis externus. As both morphological and molecular evidence strongly indicate that Kyphosus and scombroids are more closely related to other examined percomorphs than to each other , –, the derived conditions of the pharyngoclavicularis externus in each taxon are most parsimoniously considered as convergent. Therefore, the presence of an intermediate aponeurosis on the pharyngoclavicularis externus may constitute an additional synapomorphy for the Scombroidei sensu Johnson , with a reversal in Trichiuridae and a secondary change in the position of the aponeurosis occurring in Istiophoridae (Fig. 5F).
MZUSP 37378. Left lateral view; gill rakers and left rectus communis removed.
Mugiliforms and most atherinomorphs have a pharyngoclavicularis externus divided into an anterior and a posterior section (Fig. 12; see character 11). In Mugil only the anterior section of this muscle has an intermediate aponeurosis. This condition is thus not homologous to that of scombroids.
For more than a century anatomists have recognized the Stromateiformes as a natural group, mainly because of their common possession of a toothed saccular outgrowth located just posterior to the last branchial arch (Fig. 15) , –. Such a structure, called a pharyngeal sac, is unparalleled within the Actinopterygii and constitutes the most notable anatomical feature of stromateiforms. Amarsipus carlsbergi lacks a pharyngeal sac, but Haedrich  included this species within the Stromateiformes (Fig. 5G) because of other supposedly apomorphic conditions shared with remaining stromateiforms. The alignment of amarsipids with the other stromateiforms has been questioned with the demonstration that all but one character used to group these taxa have wider distributions among percomorphs , . Molecular phylogenies, strangely, split non-amarsipid stromateiforms into several separate clades that appear intercalated with bramids, caristiids, chiasmodontids, icosteids, and several scombroid lineages (Scombroidei is also resolved as polyphyletic) , , .
MZUSP 112356. Ventral view; right gill rakers, ventral part of the rectus communis, and pharyngoclavicularis externus removed; right pharyngoclavicularis internus cross-sectioned.
- 14. We confirm the presence of a toothed pharyngeal sac as a remarkable specialization unique to non-amarsipid stromateiforms (Fig. 15). This sac is covered externally by a greatly expanded sphincter oesophagi, which presents several differentiated fiber bundles. The contraction of this specialized muscle provides the primary force for the food processing that takes place in the pharyngeal sac. In anatomical descriptions and illustrations of Springer and Johnson , the stromateiform pharyngeal sac is referred to as an “epibranchial organ” [EO]. Depending on the phylogenetic allocation of the Amarsipidae in the morphology-based phylogenies of the Stromateiformes, the presence of a pharyngeal sac and its associated specialized sphincter oesophagi is optimized as either a synapomorphy for the clade formed by all non-amarsipid stromateiforms  (Fig. 5G) or a synapomorphy for the whole order, with a secondary reversal in Amarsipus . All molecular analyses including a broader taxonomic sampling of stromateiforms and their possible closer relatives ,  yielded topologies that necessarily lead to the unlikely hypothesis that the pharyngeal sac and its associated dental and muscular specializations evolved three times independently. To date, Amarsipus has not been included in any molecular phylogeny.
- 15. The bramid Brama and the non-amarsipid stromateiforms have a pharyngoclavicularis internus inserting onto a sagittal raphe shared with its antimere (Fig. 15). This raphe is dorsally continuous with the sagittal raphe of transversus ventralis V, which extends anteriorly to attach to the anterior tip of ceratobranchial 5. An insertional raphe for the pharyngoclavicularis internus was not found elsewhere among the examined acanthopterygians (Figs. 3, 6, 7, 8, 9, 10, 11, 12). Though molecular analyses allocated bramids and stromateiforms broadly into the same large monophyletic lineage , , , those two taxa were hypothesized as more closely aligned to other percomorphs lacking a sagittal raphe for the pharyngoclavicularis internus. If such scheme is accepted, then bramids and stromateiforms independently evolved this raphe, which thus might be interpreted as additional evidence for the monophyly of the non-amarsipid stromateiforms (Fig. 5G) , .
Modifications of uncertain systematic value
Some characters of the infrabranchial musculature apparently exhibit a mosaic distribution that is not immediately compatible with any known previous hypothesis of phylogenetic relationships among the major percomorph groups. These characters should not be seen as phylogenetically uncorrelated in an absolute sense, or as “noise” of lesser comparative relevance. In any morphological complex, different characters exhibit different degrees of homoplasy and this also applies to the infrabranchial musculature. The characters in this section may be diagnostic to less inclusive groups of percomorphs than those discussed in the preceding sections and/or indicate the existence of clades that have not been previously proposed in the literature. However, a correct interpretation of the phylogenetic significance of these characters will require broader phylogenetic analyses incorporating both additional taxa and characters from other body systems. In order to make these data available for future studies, characters of the infrabranchial musculature with undetermined phylogenetic signal are listed below.
- 16. Percomorphs primitively lack rectus ventralis I (Figs. 3, 6, 7, 9, 10, 11, 12, 14, 15). Among examined taxa, rectus ventralis I is present in Abudefduf, Acanthurus, Anabas, Antigonia (Fig. 13), Badis, Capros, Centrodraco, Centropyge, Chaetodon, Dactylopterus, Drepane, Elassoma, Gobiesox, Luvarus, Mastacembelus, Parachanna, Pholidichthys, Scartella (Fig. 8), Stephanolepis, Synchiropus (Fig. 10), Thalassoma, and Triacanthus.
- 17. In the plesiomorphic condition for percomorphs, obliquus ventralis I originates solely from hypobranchial 1 (Figs. 3, 6, 7, 8, 9, 10, 11). In Exocoetus, Fundulus (Fig. 12), Gasterosteus, Odontobutis, Parexocoetus, and Pungitius the origin of that muscle involves both hypobranchial 1 and basibranchial 1.
- 18. The batrachoidiforms Thalassophryne and Triathalassothia lack obliquus ventralis II. In Antennarius, Batrachoides and Lophius this muscle is vestigial. All remaining acanthopterygians exhibit a well-developed obliquus ventralis II muscle (Figs. 3, 4, 6, 7, 8, 9, 10, 11, 12, 15).
- 19. In the primitive percomorph condition, obliquus ventralis II is an undivided muscle that arises solely from hypobranchial 2 and inserts on ceratobranchial 2 (Figs. 3, 7, 8, 9, 10, 15). Atherinomorphs, gasterosteiforms , and the labrid Thalassoma have this undivided muscle expanded medially at its origin, that includes both hypobranchial 2 and basibranchial 2+3 (Fig. 12). The same two bones also serve as origin for obliquus ventralis II in gobiiforms. However, in gobiiforms the muscle is completely subdivided into an anterolateral section that originates solely from hypobranchial 2, and a posteromedial section, which arises solely from basibranchial 2+3 (character 8; Fig. 11).
- 20. Percomorphs primitively have obliquus ventralis II originating from hypobranchial 2 only (Figs. 3, 7, 8, 9, 10, 15). In Antigonia, one of the two examined specimens of Capros, triacanthids, balistoids, drepaneids, ephippidids, scatophagids, most acanthuriforms (except Luvarus and Siganus), chaetodontids, pomacentrids, and pristolepidids, obliquus ventralis II has an anterior projection that attaches to the sagittal elements of the anteriormost pharyngeal arches (urohyal, basihyal, or basibranchial 1) (Fig. 13). Some authors designate this anterior projection as rectus ventralis II –, but in our opinion such a designation might be confusing in taxa having this muscle projection completely continuous with the remainder of obliquus ventralis II. A so-called rectus ventralis II is also present in a few internested subgroups within the Blenniiformes (Fig. 8), Gasterosteiformes , and Gobiesociformes. Nevertheless, in these three orders, rectus ventralis II alternatively inserts on the paired ventrolateral elements of the first branchial arch (hypobranchial 1 and/or ceratobranchial 1).
- 21. Transversus ventralis IV may exhibit a sagittal raphe either along its entire anteroposterior extent (batrachoidiforms, zeiforms, Stephanolepis, and Triacanthus; Fig. 7) or the posterior portion of the muscle only (Apogon, Ctenopoma). In most examined taxa, however, transversus ventralis IV completely lacks a sagittal raphe (Figs. 3, 4, 6, 7, 8, 10–12, 15).
- 22. In most examined acanthopterygians transversus ventralis V has a sagittal raphe across its entire anteroposterior extent (Figs. 3, 15). Amniataba, Bathygobius (Fig. 11), Batrachoides, Bembrops, Centropomus, Chaetodon, Cynoscion, Graus, Dormitator, Lutjanus, Nototheniops, Odontobutis, Parachanna, Parexocoetus, Rhyacichthys, Scartella (Fig. 8), Scorpis, Thalassophryne, and Triathalassothia have a sagittal raphe restricted to only the posterior portion of the muscle. Transversus ventralis V completely lacks a sagittal raphe in Acanthurus, Anabas, Antennarius, Astroscopus, Atherinella, Badis, Centrodraco, Centropyge, Chaunax, Cichla (Fig. 9), Ctenopoma, Dactylopterus, Dules, Elassoma, Enneanectes, Gasterosteus, Girella, Gobiesox, Fundulus (Fig. 12), Hypsolebias, Kyphosus, Lophius, Lycodes, Microcanthus, Nandus, Ogcocephalus, Porichthys (Fig. 7), Pungitius, Scatophagus, Siganus, and Synchiropus (Fig. 10).
- 23. Most examined acanthopterygians have the sphincter oesophagi and transversus ventralis V completely separated from each other (Figs. 3, 6–8, 10, 11, 12, 15). These muscles are partially continuous with each other in Chaetodipterus, Cichla (Fig. 9), Drepane, Dules, Elassoma, Enneanectes, Gasterosteus, Lophius, Lycodes, Mastacembelus, Nototheniops, Ogcocephalus, Parachanna, Parexocoetus, Rhyacichthys, Synbranchus, and Triacanthus.
- 24. Primitively in percomorphs, transversus ventralis V is undifferentiated into subsections (Figs. 3, 7, 9, 15). This muscle is differentiated in an anterior and a posterior section in Anabas, Antennarius, Badis, Bathygobius (Fig. 11), Centropomus, Ctenopoma, Cynoscion, Dormitator, Enneanectes, Fundulus (Fig. 12), Hypsolebias, Lutjanus, Monodactylus, Nandus, Nototheniops, Odontobutis, Orthopristis, Pristolepis (Fig. 6), Scartella (Fig. 8), Stephanolepis, and Synchiropus (Fig. 10).
- 25. Lower euteleosts apparently have the rectus communis originating from hypobranchial 3 only , . Stiassny  and Lauder  considered the shift in origin of this muscle to the urohyal as a ctenosquamate synapomorphy. The vast majority of the examined acanthomorphs indeed have a rectus communis originating solely from the urohyal. In Brama, Holocentrus, Peprilus (Fig. 15), Prionotus, and Psettodes the ventral portion of the muscle arises from the urohyal, but a dorsal set of fibers attach to hypobranchial 3. However, Paralichthys, Poromitra, Trichiurus, Zenion, and Zenopsis have a rectus communis originating from hypobranchial 3 only, such as in the putative plesiomorphic condition for the Euteleosteomorpha. Lophiiforms uniquely have a rectus communis arising from the dorsal hypohyal (character 4).
- 26. The lophiiforms Antennarius, Lophius, and Ogcocephalus and the batrachoidiforms Batrachoides and Thalassophryne have the rectus communis inserting musculously on ceratobranchial 5. In all other examined acanthopterygians, this insertion is completely tendinous (Figs. 3, 6, 7, 10, 13, 15).
- 27. Most examined acanthopterygians have the insertion of the pharyngoclavicularis internus positioned anteromedial to the insertion of the pharyngoclavicularis externus (Figs. 3, 9, 11, 12, 14, 15). As a consequence, the former muscle passes medial to the latter. In examined anabantomorphs (Fig. 6), batrachoidiforms (Fig. 7), lophiiforms, uranoscopids, blenniiforms (Fig. 8), callionymoids (Fig. 10), gobiesocoids, mastacembeloids, synbranchoids, elassomatiforms, Odontobutis, and Siganus, the insertion of the pharyngoclavicularis internus is posterior to the insertion of the pharyngoclavicularis externus. Dormitator, Gasterosteus, and Lycodes alternatively have a pharyngoclavicularis internus inserting posterolateral to the pharyngoclavicularis externus.
- 28. In the primitive percomorph condition, the pharyngoclavicularis externus inserts musculously on ceratobranchial 5 (Figs. 3, 6, 7, 8, 9, 10, 11, 12, 14, 15). In Antigonia, acanthurids, and most tetraodontiforms, the insertion of this muscle is entirely mediated by tendon (Fig. 13) –.
- 29. The caproiform Antigonia and non-tetraodontoid tetraodontiforms have a fully differentiated sternobranchialis, a muscle derived from the separation of the dorsomedial portion of the sternohyoideus (Fig. 13) . Elsewhere in percomorphs, only gobiiforms also have a well-differentiated sternobranchialis (Fig. 11). However, the gobiiform sternobranchialis differs from that of Antigonia and tetraodontiforms in many features (see character 9) and analyses of different data sets (osteology, myology, and DNA) indicate that gobiiforms are not closely related to caproiforms or tetraodontiforms , , , , , , . Therefore, the sternobranchialis of gobiiforms apparently is not homologous to the sternobranchialis of Antigonia and tetraodontiforms, which is characterized by having a bulged origin and strongly flattened insertion, a nearly vertical orientation, antimeres much closer to one another, and an insertional aponeurosis splitting into two or more tendons that attach to several bones of the branchial and/or hyoid arches (Fig. 13).
The infrabranchial musculature has been unexplored in most fish groups, and the present study shows that many modifications in this system may have important implications to the higher-level phylogeny of percomorphs. Data from the infrabranchial muscles notably support several hypotheses of percomorph relationships advanced by preceding anatomists on the basis of distinct data sources (mainly osteology). Interestingly, most of these relationships were never recovered in recent molecular phylogenies, such as the sister-group relationship between the Batrachoidiformes and Lophiiformes –, Callionymoidei and Gobiesocoidei , , –, and Mugiliformes and Atherinomorphae , ; the monophyly of the Scombroidei  and the non-amarsipid Stromateiformes ; the inclusion of the Pholidichthyiformes within the Labriformes , ; and the basal allocation of the Rhyacichthyidae and Odontobutidae as successive sister-groups to remaining Gobiiformes – (Fig. 5). The existence of independent anatomical evidence repeatedly favoring a same set of hypotheses discordant with the molecular topologies clearly indicates that the interrelationships of most major percomorph groups is still far from a satisfactory resolution.
The lack of basic comparative data for several anatomical systems is a major problem that hampers the production of morphology-based phylogenies of ray-finned fishes. The primary homologies (or hypotheses of homologies) of several structures are often unclear, especially among major actinopterygian groups, and the production of comprehensive, detailed studies of comparative anatomy [e.g. 10,48,114–116] is crucial to mitigate this sort of problem. The present study is an attempt to provide basic and detailed knowledge on the percomorph infrabranchial musculature, so that this anatomical system may be more satisfactorily incorporated into larger data matrices of future phylogenetic analyses. Denser sampling of percomorphs and incorporation of characters from other systems may, and likely will, reveal important phylogenetic signal in all characters herein discussed, including those which now seem to display mosaic distributions.
Thanks to Richard P. Vari (USNM), Flávio A. Bockmann (LIRP), Ricardo M. C. Castro (LIRP), Karsten Hartel (MCZ), Ronald Fritzsche (HSU), and H. J. Walker (SIO) for allowing the dissection of specimens critical to this study. Masanori Nakae (National Museum of Nature and Science, Japan) kindly provided important information on the infrabranchial musculature of some scombroids. Flávio A. Bockmann generously granted access to his stereomicroscopic photographic equipment and Luciana F. Tosin edited several photographs.
Conceived and designed the experiments: AD MCCP. Performed the experiments: AD GDJ. Analyzed the data: AD GDJ. Contributed reagents/materials/analysis tools: AD MCCP GDJ. Wrote the paper: AD MCCP GDJ.
- 1. Wiley EO, Johnson GD (2010) A teleost classification based on monophyletic groups. In: Nelson JS, Schultze H-P, Wilson MVH, editors. Origin and Phylogenetic Interrelationships of Teleosts. München: Verlag Dr. Friedrich Pfeil. pp. 123–182.
- 2. Eschmeyer WN, Fong JD (2014) Species of fishes by family/subfamily. Available: http://research.calacademy.org/research/ichthyology/catalog/SpeciesByFamily.asp. Accessed 05 March 2014.
- 3. IUCN (2012) The IUCN Red List of Threatened Species, v. 2012.1, Sumary Statictics. Available: http://www.iucnredlist.org/about/summary-statistics. Accessed 30 March 2013.
- 4. Chen W, Bonillo C, Lecointre G (2003) Repeatability of clades as a criterion of reliability: a case study for molecular phylogeny of Acanthomorpha (Teleostei) with larger number of taxa. Mol Phylogenet Evol 26: 262–288.
- 5. Dettai A, Lecointre G (2005) Further support for the clades obtained by multiple molecular phylogenies in the acanthomorph bush. C R Biol 328: 674–689.
- 6. Miya M, Kawaguchi A, Nishida M (2001) Mitogenomic exploration of higher teleostean phylogenies: a case study for moderate-scale evolutionary genomics with 38 newly determined complete mitochondrial DNA sequences. Mol Phylogenet Evol 18: 1993–2009.
- 7. Miya M, Takeshima H, Endo H, Ishiguro NB, Inoue JG, et al. (2003) Major patterns of higher teleostean phylogenies: a new perspective based on 100 complete mitochondrial DNA sequences. Mol Phylogenet Evol 26: 121–138.
- 8. Near TJ, Dornburg A, Eytan RI, Keck BP, Smith WL, et al. (2013) Phylogeny and tempo of diversification in the superradiation of spiny-rayed fishes. Proc Natl Acad Sci U S A 110: 12738–12743.
- 9. Betancur-R R, Broughton RE, Wiley EO, Carpenter K, López JA, et al.. (2013) The tree of life and a new classification of bony fishes. PLOS Currents Tree of Life 2013 Apr 18 [last modified: 2013 Jun 3]. Edition 1.
- 10. Springer VG, Johnson GD (2004) Study of the dorsal gill-arch musculature of teleostome fishes, with special reference to the Actinopterygii. Bull Biol Soc Wash 11: vi+1–260+pl. 001–205.
- 11. Nelson GJ (1969) Gill arches and the phylogeny of fishes, with notes on the classification of vertebrates. Bull Am Mus Nat Hist 141: 475–552.
- 12. Nelson GJ (1967) Branchial muscles in some generalized teleostean fishes. Acta Zool (Stockh) 48: 277–288.
- 13. Nelson GJ (1967) Gill arches of some teleostean fishes of the families Girellidae, Pomacentridae, Embiotocidae Labridae and Scaridae. J Nat Hist 1: 289–293.
- 14. Nelson GJ (1967) Branchial muscles in representatives of five eel families. Pac Sci 21: 348–363.
- 15. Nelson GJ (1966) Gill arches of teleostean fishes of the order Anguilliformes. Pac Sci 20: 391–408.
- 16. Nelson GJ (1970) Gill arches of some teleostean fishes of the families Salangidae and Argentinidae. Jpn J Ichthyol 17: 61–66.
- 17. Nelson G (1968) Gill arches of teleostean fishes of the division Osteoglossomorpha. Zool J Linn Soc 47: 261–277.
- 18. Nelson G (1967) Epibranchial organs in lower teleostean fishes. J Zool (Lond) 153: 71–89.
- 19. Stiassny MLJ, Jensen JS (1987) Labroid intrarelationships revisited: morphological complexity, key innovations, and the study of comparative diversity. Bull Mus Comp Zool 151: 269–319.
- 20. Allis EP Jr (1903) The skull, and the cranial and first spinal muscles and nerves in Scomber scomber. J Morphol 18: 45–328.
- 21. Anker GC (1978) The morphology of the head-muscles of a generalized Haplochromis species: H. elegans Trewavas 1933 (Pisces, Cichlidae). Neth J Zool 28: 234–271.
- 22. Howes GJ (1976) The cranial musculature and taxonomy of characoid fishes of the tribes Cynodontini and Characini. Bull Br Mus (Nat Hist) Zool 29: 203–248.
- 23. Lauder GV, Liem KF (1980) The feeding mechanism and cephalic myology of Salvelinus fontinalis: form, function, and evolutionary significance. In: Balon EK, editor. Charrs: Salmonid fishes of the genus Salvelinus. The Hague: Dr. W. Junk bv Publishers. 365–390.
- 24. Stiassny MLJ (1990) Notes on the anatomy and relationships of the bedotiid fishes of Madagascar, with a taxonomic revision of the genus Rheocles (Atherinomorpha: Bedotiidae). Am Mus Novit 2979: 1–33.
- 25. Takata Y, Sasaki K (2005) Branchial structures in the Gasterosteiformes, with special reference to myology and phylogenetic implications. Ichthyol Res 52: 33–49.
- 26. Anker GC (1974) Morphology and kinetics of the head of the stickleback, Gasterosteus aculeatus. Trans Zool Soc London 32: 311–416.
- 27. Vandewalle P, Parmentier E, Poulicek M (1998) Bussers (1998) Chardon M (1998) Distinctive anatomical features of the branchial basket in four Carapidae species (Ophidiiformi, Paracanthopterygii). Eur J Morphol 36: 153–164.
- 28. Lauder GV (1983) Functional design and evolution of the pharyngeal jaw apparatus in euteleostean fishes. Zool J Linn Soc 77: 1–38.
- 29. Elshoud-Oldenhave MJW (1979) Prey capture in the pike-perch, Stizostedion lucioperca (Teleostei, Percidae): a structural and functional analysis. Zoomorphology (Berl) 93: 1–32.
- 30. Vandewalle P, Saintin P, Chardon M (1995) Structures and movements of the buccal and pharyngeal jaws in relation to feeding in Diplodus sargus. J Fish Biol 46: 623–656.
- 31. Eschmeyer WN (2014) Catalog of Fishes electronic version. Available: http://research.calacademy.org/redirect?url=http://researcharchive.calacademy.org/research/Ichthyology/catalog/fishcatmain.asp. Accessed 05 March 2014.
- 32. Datovo A, Bockmann FA (2010) Dorsolateral head muscles of the catfish families Nematogenyidae and Trichomycteridae (Siluriformes: Loricarioidei): comparative anatomy and phylogenetic analysis. Neotrop Ichthyol 8: 193–246.
- 33. Winterbottom R (1974) A descriptive synonymy of the striated muscles of the Teleostei. Proc Acad Nat Sci Phila 125: 225–317.
- 34. Hadley A (2009) CombineZP. GNU public license software.
- 35. Stiassny MLJ (1992) Atavisms, phylogenetic character reversals, and the origin of evolutionary novelties. Neth J Zool 42: 260–276.
- 36. Stiassny MLJ (1996) Basal ctenosquamate relationships and the interrelationships of the myctophiform (scopelomorph) fishes. In: Stiassny MLJ, Parenti LR, Johnson GD, editors. Interrelationships of Fishes. San Diego: Academic Press. 333–354.
- 37. Shinohara G (1994) Comparative morphology and phylogeny of the suborder Hexagrammoidei and related taxa (Pisces: Scorpaeniformes). Mem Fac Fish Hokkaido Univ 41: 1–97.
- 38. Yabe M, Uyeno T (1996) Anatomical description of Normanichthys crockeri (Scorpaeniformes, incertae sedis: family Normanichthyidae). Bull Mar Sci 58: 494–510.
- 39. Imamura H (1996) Phylogeny of the family Platycephalidae and related taxa (Pisces: Scorpaeniformes). Spec Divers 1: 123–233.
- 40. Ishida M (1994) Phylogeny of the suborder Scorpaenoidei (Pisces: Scorpaeniformes). Bull Nansei Natl Fish Res Inst 27: 1–112.
- 41. Borden WC, Coburn MM (2008) Striated muscles of the black basses (Micropterus, Centrarchidae): myological stasis in a generalized group of percomorph fishes. Bull Fla Mus Nat Hist 47: 109–136.
- 42. Aguilera O (1987) Musculatura de los arcos branquiales, arco hioideo y operculo en peces Sciaenidae del Atlantico Oeste y rios de America. Rev Bras Zool 4: 89–103.
- 43. Sasaki K (1989) Phylogeny of the family Sciaenidae, with notes on its zoogeography (Teleostei, Perciformes). Mem Fac Fish Hokkaido Univ 36: 1–137.
- 44. Gosline WA (1968) The suborders of perciform fishes. Proc U S Natl Mus 124: 1–78.
- 45. Gosline WA (1971) Functional morphology and classification of teleostean fishes. Honolulu: University Press of Hawaii. 208 p.
- 46. Gosline WA (1985) A possible relationship between aspects of dentition and feeding in the centrarchid and anabantoid fishes. Environ Biol Fishes 12: 161–168.
- 47. Rosen DE, Patterson C (1969) The structure and relationships of the paracanthopterygian fishes. Bull Am Mus Nat Hist 141: 361–474+ pl. 352–378.
- 48. Rosen DE, Patterson C (1990) On Muller's and Cuvier's concepts of pharyngognath and labyrinth fishes and the classification of percomorph fishes, with an atlas of percomorph dorsal gill arches. Am Mus Novit 2983: 1–57.
- 49. Britz R (2003) Suborder Anabantoidei (Labyrinth fishes). In: Hutchins M, Thoney DA, Loiselle PV, Schlager N, editors. Grzimek’s Animal Life Encyclopedia, Vol 5, Fishes II. Farmington Hills: Gale Group. 427–436.
- 50. Betancur-R R, Wiley EO, Miya M, Lecointre G, Bailly N, et al.. (2013) New and revised classification of bony fishes, Version 2. Available: http://www.deepfin.org/Classification_v2.htm. Accessed 05 March 2014.
- 51. Regan CT (1909) The classification of teleostean fishes. Ann Mag Nat Hist 8: 75–86.
- 52. Regan CT (1912) The classification of the teleostean fishes of the order Pediculati. Ann Mag Nat Hist 8: 277–289.
- 53. Lauder GV, Liem KF (1983) The evolution and interrelationships of the actinopterygian fishes. Bull Mus Comp Zool 150: 95–195.
- 54. Patterson C, Rosen DE (1989) The Paracanthopterygii revisited: order and disorder. In: Cohen DM, editor. Papers on the Systematics of Gadiform Fishes. Los Angeles: Natural History Museum of Los Angeles County.
- 55. Miya M, Satoh TP, Nishida M (2005) The phylogenetic position of toadfishes (order Batrachoidiformes) in the higher ray-finned fish as inferred from partitioned Bayesian analysis of 102 whole mitochondrial genome sequences. Biol J Linn Soc 85: 289–306.
- 56. Near TJ, Eytan RI, Dornburg A, Kuhn KL, Moore JA, et al. (2012) Resolution of ray-finned fish phylogeny and timing of diversification. Proc Natl Acad Sci U S A 109: 13698–13703.
- 57. Greenfield DW, Winterbottom R, Collette BB (2008) Review of the Toadfish Genera (Teleostei: Batrachoididae). Proc Calif Acad Sci 59: 665–710.
- 58. Miya M, Pietsch TW, Orr JW, Arnold RJ, Satoh TP, et al. (2010) Evolutionary history of anglerfishes (Teleostei: Lophiiformes): a mitogenomic perspective. BMC Evol Biol 10: 1–27.
- 59. Pietsch TW, Orr JW (2007) Phylogenetic relationships of deep-sea anglerfishes of the suborder Ceratioidei (Teleostei: Lophiiformes) based on morphology. Copeia 2007: 1–34.
- 60. Baldwin C (2013) The phylogenetic significance of colour patterns in marine teleost larvae. Zool J Linn Soc 168: 496–563.
- 61. Chanet B, Guintard C, Betti E, Gallut C, Dettai A, et al. (2013) Evidence for a close phylogenetic relationship between the teleost orders Tetraodontiformes and Lophiiformes based on an analysis of soft anatomy. Cybium 37: 179–198.
- 62. Pietsch TW (1981) The osteology and relationships of the anglerfish genus Tetrabrachium with comments on lophiiform classification. Fish Bull (Seattle) 79: 387–419.
- 63. Pietsch TW, Grobecker DB (1987) Frogfishes of the world. Systematics, zoogeography & behavioral ecology. Stanford: Stanford University Press. 420 p.
- 64. Springer VG, Orrel TM (2004) Appendix: phylogenetic analysis of 147 families of acanthomorph fishes based primarily on dorsal gill-arch muscles and skeleton – In: Springer, V. G. & Johnson, G. D. Study of the dorsal gill-arch musculature of teleostome fishes, with special reference to the Actinopterygii. Bull Biol Soc Wash 11: 236–260.
- 65. Greenwood PH, Rosen DE, Weitzman SH, Myers GS (1966) Phyletic studies of teleostean fishes, with a provisional classification of living forms. Bull Am Mus Nat Hist 131: 339–456.
- 66. Gosline WA (1970) A reinterpretation of the teleostean fish order Gobiesociformes. Proc Calif Acad Sci 38: 363–382.
- 67. Johnson GD, Patterson C (1993) Percomorph phylogeny: a survey of acanthomorphs and a new proposal. Bull Mar Sci 52: 554–626.
- 68. Winterbottom R (1993) Search for the gobioid sister group (Actinopterygii: Percomorpha). Bull Mar Sci 52: 395–414.
- 69. Li B, Dettaï A, Cruaud C, Couloux A, Desoutter-Meniger M, et al. (2009) RNF213, a new nuclear marker for acanthomorph phylogeny. Mol Phylogenet Evol 50: 345–363.
- 70. Dietz PA (1914) Beiträge zur Kenntnis der Kiefer- und Kiemenbogenmuskulatur der Teleostier. I, Die Kiefer und Kiemenbogenmuskeln der Acanthopterygier. Mitt Zool Sta Neapel 22: 99–162.
- 71. Borden WC (1999) Comparative myology of the unicornfishes, Naso (Acanthuridae, Percomorpha), with implications for phylogenetic analysis. J Morphol 239: 191–224.
- 72. Winterbottom R (1993) Myological evidence for the phylogeny of recent genera of surgeonfishes (Percomorpha, Acanthuridae), with comments on the Acanthuroidei. Copeia 1993: 21–39.
- 73. Winterbottom R (1974) The familial phylogeny of the Tetraodontiformes (Acanthopterygii: Pisces) as evidenced by their comparative myology. Smithson Contrib Zool 155: 1–201.
- 74. Imamura H, Shirai S, Yabe M (2005) Phylogenetic position of the family Trichodontidae (Teleostei: Perciformes), with a revised classification of the perciform suborder Cottoidei. Ichthyol Res 52: 264–274.
- 75. Yabe M (1985) Comparative osteology and myology of the superfamily Cottoidea (Pisces: Scorpaeniformes), and its phylogenetic classification. Mem Fac Fish Hokkaido Univ 32: 1–130.
- 76. Hoese DF, Gill AC (1993) Phylogenetic relationships of eleotridid fishes (Perciformes: Gobioidei). Bull Mar Sci 52: 415–440.
- 77. Johnson GD, Brothers EB (1993) Schindleria, a paedomorphic goby (Teleostei: Gobioidei). Bull Mar Sci 52: 441–471.
- 78. Springer VG (1983) Tyson belos, new genus and species of Western Pacific fish (Gobiidae, Xenisthminae), with discussions of gobioid osteology and classification. Smithson Contrib Zool 390: 1–40.
- 79. Chakrabarty P, Davis MP, Sparks JS (2012) The first record of a trans-oceanic sister-group relationship between obligate vertebrate troglobites. PLOS ONE 7: e44083.
- 80. Thacker CE (2009) Phylogeny of Gobioidei and placement within Acanthomorpha, with a new classification and investigation of diversification and character evolution. Copeia 2009: 93–104.
- 81. Stiassny MLJ (1993) What are grey mullets? Bull Mar Sci 52: 197–219.
- 82. Thacker CE (2003) Molecular phylogeny of the gobioid fishes (Teleostei: Perciformes: Gobioidei). Mol Phylogenet Evol 26: 354–368.
- 83. Gobalet KW (1989) Morphology of the parrotfish pharyngeal jaw apparatus. Am Zool 29: 319–331.
- 84. Stiassny MLJ (1987) Cichlid familial intrarelationships and the placement of the neotropical genus Cichla (Perciformes, Labroidei). J Nat Hist 21: 1311–1331.
- 85. Clements KD, Bellwood DR (1988) A comparison of the feeding mechanisms of two herbivorous labroid fishes, the temperate Odax pullus and the tropical Scarus rubroviolaceus. Aust J Mar Freshw Res 39: 87–107.
- 86. Britz R (2006) Book review: Springer, V.G. & Johnson, G.D. 2004. Study of the dorsal gill-arch musculature of teleostome fishes, with special reference to the Actinopterygii. Bulletin of the Biological Society of Washington, 11: p. i-vi +1–260, pl. 1–205. Copeia 2006: 323–330.
- 87. Springer VG, Freihofer WC (1976) Study of the monotypic fish family Pholidichthyidae (Perciformes). Smithson Contrib Zool 216: 1–43.
- 88. Nelson JS (2006) Fishes of the World. New York: John Wiley & Sons Inc. 601 p.
- 89. Kaufman LS, Liem KF (1982) Fishes of the suborder Labroidei (Pisces: Perciformes): phylogeny, ecology, and evolutionary significance. Breviora 472: 1–19.
- 90. Johnson GD (1993) Percomorph phylogeny: progress and problems. Bull Mar Sci 52: 3–28.
- 91. Wainwright PC, Smith WL, Price SA, Tang KL, Sparks JS, et al. (2012) The evolution of pharyngognathy: a phylogenetic and functional appraisal of the pharyngeal jaw key innovation in labroid fishes and beyond. Syst Biol 61: 1001–1027.
- 92. Parenti L (1993) Relationships of atherinomorph fishes (Teleostei). Bull Mar Sci 52: 170–196.
- 93. Parenti L (2005) The phylogeny of atherinomorphs: evolution of a novel fish reproductive system. In: Uribe MC, Grier HJ, editors. Viviparous fishes. Homstead: New Life Publ. 13–30.
- 94. Rosen DE (1964) The relationships and taxonomic position of the halfbeaks, killifishes, silversides, and their relatives. Bull Am Mus Nat Hist 127: 217–268, pl. 214–215.
- 95. Rosen DE, Parenti L (1981) Relationships of Oryzias, and the groups of atherinomorph fishes. Am Mus Novit 2719: 1–25.
- 96. Setiamarga DHE, Miya M, Yamanoue Y, Mabuchi K, Satoh TP, et al. (2008) Interrelationships of Atherinomorpha (medakas, flyingfishes, killifishes, silversides, and their relatives): the first evidence based on whole mitogenome sequences. Mol Phylogenet Evol 49: 598–605.
- 97. Collette BB, Potthoff T, Richards WJ, Ueyanagi S, Russo JL, et al.. (1984) Scombroidei: development and relationships. In: Moser HG, Richards WJ, Cohen DM, Fahay MP, Kendall AW, Jr et al.., editors. Ontogeny and Systematics of Fishes. Lawrence: American Society of Ichthyologists and Herpetologists. 591–620.
- 98. Johnson GD (1986) Scombroid phylogeny: an alternative hypothesis. Bull Mar Sci 39: 1–41.
- 99. Carpenter KE, Collette BB, Russo JL (1995) Unstable and stable classifications of scombroid fishes. Bull Mar Sci 56: 379–405.
- 100. Jordan DS (1905) A guide to the study of fishes. Vol. 2. New York: Henry Holt and Co. xxii+599 p.
- 101. Regan CT (1909) On the anatomy and classification of the scombroid fishes. J Nat Hist 3: 66–75.
- 102. Tyler JC, Johnson GD, Nakamura I, Collette BB (1989) Morphology of Luvarus imperialis (Luvaridae), with a phylogenetic analysis of the Acanthuroidei (Pisces). Smithson Contrib Zool 485: 1–78.
- 103. Starks EC (1909) The scombroid fishes. Science 30: 572–574.
- 104. Yagishita N, Miya M, Yamanoue Y, Shirai SM, Nakayama K, et al. (2009) Mitogenomic evaluation of the unique facial nerve pattern as a phylogenetic marker within the percifom fishes (Teleostei: Percomorpha). Mol Phylogenet Evol 53: 258–266.
- 105. De Schepper N, Van Vassenbergh S, Adriaens D (2008) Morphology of the jaw system in trichiurids: trade-offs between mouth closing and biting performance. Zool J Linn Soc 152: 717–736.
- 106. Freihofer WC (1963) Patterns of the ramus lateralis acessorius and their phylogenetic significance in teleostean fishes. Stanford Ichthyol Bull 8: 79–189.
- 107. Johnson GD, Fritzsche RA (1989) Graus nigra, a omnivorous girellid, with a comparative osteology and comments on relationships of the Gireliidae (Pisces: Perciformes). Proc Acad Nat Sci Phila 141: 1–27.
- 108. Miya M, Friedman M, Satoh TP, Takeshima H, Sado T, et al. (2013) Evolutionary origin of the Scombridae (tunas and mackerels): members of a Paleogene adaptive radiation with 14 other pelagic fish families. PLOS ONE 8: e73535.
- 109. Haedrich RL (1967) The stromateoid fishes: systematics and a classification. Bull Mus Comp Zool 135: 31–139.
- 110. Haedrich RL (1969) A new family of aberrant stromateoid fishes from the equatorial Indo-Pacific. Dana-Rep 76: 1–14.
- 111. Horn MH (1984) Stromateoidei: Development and relationships. In: Moser HG, Richards WJ, Cohen DM, Fahay MP, Kendall AW, Jr et al.., editors. Ontogeny and Systematics of Fishes. Lawrence: American Society of Ichthyologists and Herpetologist. 620–636.
- 112. Regan CT (1902) A revision of the fishes of the family Stromateidae. Ann Mag Nat Hist Ser. 7 10: 115–131.
- 113. Doiuchi R, Sato T, Nakabo T (2004) Phylogenetic relationships of the stromateoid fishes (Perciformes). Ichthyol Res 51: 202–212.
- 114. Datovo A, Vari RP (2013) The jaw adductor muscle complex in teleostean fishes: evolution, homologies and revised nomenclature (Osteichthyes: Actinopterygii). PLOS ONE 8: e60846.
- 115. Datovo A, Vari RP (2014) The adductor mandibulae muscle complex in lower teleostean fishes (Osteichthyes: Actinopterygii): comparative anatomy, synonymy, and phylogenetic implications. Zool J Linn Soc 171: 554–622.
- 116. Patterson C, Johnson GD (1995) The intermuscular bones and ligaments of teleostean fishes. Smithson Contrib Zool 559: 1–83.
- 117. Kullander SO, Britz R (2002) Revision of the family Badidae (Teleostei: Perciformes), with description of a new genus and ten new species. Ichthyol Explor Freshwat 13: 295–372.