The book represents contributions to the symposium “Origin and phylogenetic interrelationships of teleosts” sponsored by the American Society of Ichthyologists and Herpetologists (ASIH) and organized by the three editors of this volume and held at the Society’s annual meeting in St. Louis, Missouri, on 14 July 2007. At the same meeting, Gloria Arratia was honored with the Robert H. Gibbs, Jr. Memorial Award, 2007, for her outstanding contributions to systematic ichthyology. The volume presents the current state of phylogenetic knowledge of the origin of teleosts and the interrelationships of teleost groups, both key issues in fish systematics, based on both morphological (of extant and fossil taxa) and molecular evidence. The many contributors to the volume present and evaluate progress in studying both characters and taxa and in establishing databases (morphological and molecular) that will be of use in future.

Preface 7-9
Acknowledgments 9
Joseph S. NELSON:
Gloria Arratia’s contribution to our understanding of lower teleostean phylogeny and classification 11-36
J. Ralph NURSALL:
The case for pycnodont fishes as the fossil sister-group of teleosts 37-60
Richard E. BROUGHTON:
Phylogeny of teleosts based on mitochondrial genome sequences 61-76
Ralf BRITZ and G. David JOHNSON:
Occipito-vertebral fusion in actinopterygians: conjecture, myth and reality. Part 1: Non-teleosts 77-93
G. David JOHNSON and Ralf BRITZ:
Occipito-vertebral fusion in actinopterygians: conjecture, myth and reality. Part 2: Teleosts 95-110
Lionel CAVIN:
The Late Jurassic ray-finned fish peak of diversity: biological radiation or preservational bias? 111-121
E. O. WILEY and G. David JOHNSON:
A teleost classification based on monophyletic groups 123-182
Peter L. FOREY and John G. MAISEY:
Structure and relationships of †Brannerion (Albuloidei), an Early Cretaceous teleost from Brazil 183-218
Eric J. HILTON and Ralf BRITZ:
The caudal skeleton of osteoglossomorph fishes, revisited: comparisons, homologies, and characters 219-237
ZHANG JIANG-YONG:
Validity of the osteoglossomorph genus †Asiatolepis and a revision of †Asiatolepis muroii (†Lycoptera muroii) 239-249
Mário DE PINNA and Fábio DI DARIO:
The branchial arches of the primitive clupeomorph fish, Denticeps clupeoides, and their phylogenetic implication (Clupeiformes, Denticipitidae) 251-268
Francisco José POYATO-ARIZA, Terry GRANDE and Rui DIOGO:
General overview of fossil and Recent Gonorynchiformes (Teleostei, Ostariophysi) 269-293
Kevin W. CONWAY, M. Vincent HIRT, Lei YANG, Richard L. MAYDEN and Andrew M. SIMONS:
Cypriniformes: systematics and paleontology 295-316
Maria Claudia MALABARBA and Luiz R. MALABARBA:
Biogeography of Characiformes: an evaluation of the available information of fossil and extant taxa 317-336
Dominique ADRIAENS, Jonathan N. BASKIN and Hendrik COPPENS:
Evolutionary morphology of trichomycterid catfishes: about hanging on and digging in 337-362
Jacob J. D. EGGE:
Systematics of ictalurid catfishes: a review of the evidence 363-378
Mark V. H. WILSON and Robert R. G. WILLIAMS:
Salmoniform fishes: key fossils, supertree, and possible morphological synapomorphies 379-409
Amanda BURDI and Terry GRANDE:
Morphological development of the axial skeletons of Esox lucius and Esox masquinongy (Euteleostei: Esociformes), with comparisons in developmental and mineralization rates 411-430
Matthew P. DAVIS:
Evolutionary relationships of the Aulopiformes (Euteleostei: Cyclosquamata): a molecular and total evidence approach 431-470
Irma VILA, Sergio SCOTT, Natalia LAM, Patricia ITURRA and Marco A. MÉNDEZ:
Karyological and morphological analysis of divergence among species of the killifish genus Orestias (Teleostei: Cyprinodontidae) from the southern Altiplano 471-480

Joseph S. NELSON:
Gloria Arratia’s contribution to our understanding of lower teleostean phylogeny and classification
[pp. 11-36, 7 coloured and 1 black-and-white figures, 2 tables]

Few ichthyologists have contributed as greatly to the field of lower teleost systematics as Dr. Gloria Arratia. Born in 1942 in Santiago, Chile, Gloria became interested in evolution and its processes while a high school student. This interest was further developed at university where she studied vertebrate morphology in much greater detail than was common at the time. After she graduated from the University of Chile she began to work on extant Chilean catfishes, gaining much expertise on fish osteology. She subsequently started working also on fossil fishes, beginning in the Atacama Desert. In her highly successful career she has conducted research in Chile, Argentina, Germany, Sweden, and the U.S.A. Ever since the mid-1980s, Gloria has been a leading figure in resolving many questions that has given us a better understanding of the origins and limits of the teleosts and of early teleost phylogeny. This has largely been through detailed studies of the morphology of fossil and recent fishes, allowing her to solve problems of homology and to better understand the evolution of characters and evolutionary changes within various groups. She has described many taxa, both living and fossil. She has made major contributions in understanding siluriform morphology and evolution, in understanding basal teleosts morphology and evolution, and in providing valuable insight into the problem of which is the more primitive extant teleost taxon, the osteoglossomorphs or the elopomorphs. She is involved in numerous research projects, e. g., she is currently one of the principal investigators in the Tree of Life of Cypriniformes, supported by NSF, and conducting morphological and developmental studies. Amongst her many editorial functions and administrative duties, she has served as Editor-in-Chief of the Mesozoic Fishes volumes and spearheaded the four Mesozoic Fishes conferences. In recognition of her outstanding contributions, she has received numerous awards, including the Alexander von Humboldt Prize, 1994, the Robert H. Gibbs, Jr. Memorial Award, 2007, and the Artedi Lecturer Diploma, 2008.

J. Ralph NURSALL:
The case for pycnodont fishes as the fossil sister-group of teleosts
[pp. 37-60, 9 black-and-white figures, 2 tables]

Pycnodont fishes, distinguished and named by Agassiz, comprise a large, unique, monophyletic group of considerable variety, whose systematic position has never been clearly established. Now the Superorder †Pycnodontomorpha (new), within Halecostomi, is proposed for the fossil fishes currently included in the Order †Pycnodontiformes. The new Superorder comprises at least two orders: †Pycnodontiformes (new use) and †Gyrodontiformes (new). Nineteen persistent, functional, non-trivial apomorphies are used to characterize †Pycnodontomorpha. Erection of the new Superorder is justified by the unique, congruent characteristics that distinguish the taxon. †Pycnodontomorpha persisted through an extensive period of time; there is a large number and diversity of taxa defined within the Superorder; pycnodontomorphs achieved near-global distribution. Analysis of the characteristics of the new taxon leads to the conclusion that †Pycnodontomorpha are the sister-group to Teleosteomorpha, with which they share a number of character states. The two groups also shared habitat in Late Cretaceous and Early Tertiary times.

Richard E. BROUGHTON:
Phylogeny of teleosts based on mitochondrial genome sequences
[pp. 61-76, 2 black-and-white figures, 1 table, 1 appendix]

Mitochondrial DNA sequences have long been used for molecular phylogenetic analyses; however, their ability to resolve deep diverging lineages has been mixed. Recently, mitochondrial genome sequences have been applied to many questions in fish phylogeny and systematics. Using data sets with large numbers of characters may be useful for resolving higher taxa such as families and orders. Relationships among many actinopterygian orders or other higher groups remain elusive based on morphological and limited molecular data. I used a set of all 13 mitochondrial protein coding genes from 230 mitochondrial genomes in a large-scale phylogenetic analysis of teleost fishes. The analysis included all available taxa from many basal teleost families representing all basal orders. Maximum likelihood and Bayesian analyses revealed a general structure of teleost relationships with many current hypotheses supported. However, some clades that are important for understanding teleost diversification were not recovered with strong support. Analyses revealed that searches for optimal phylogenetic trees were sensitive to nucleotide composition, taxon sampling, and outgroup selection. The resulting best phylogenetic hypothesis is discussed in the context of other recent molecular phylogenetic studies of fishes and with respect to conventionally understood teleost interrelationships.

Ralf BRITZ and G. David JOHNSON:
Occipito-vertebral fusion in actinopterygians: conjecture, myth and reality. Part 1: Non-teleosts
[pp. 77-93, 6 coloured and 1 black-and-white figures]

We revisit the century old hypothesis that the occipita of different actinopterygians are not comparable, because a varying number of vertebrae forms the occiput in the different actinopterygian groups. We applied an ontogenetic approach to this issue and also utilized the number of myosepta that attach to the back of the skull as indication for occipito-vertebral fusion. We found that there is no incorporation of vertebrae or parts thereof into the occiput in Polypterus and Erpetoichthys in development. Uniquely among actinopterygians, the first neural arch fails to develop a corresponding centrum in polypterids and shifts onto the occipital bone in later development, creating the false impression of an accessory neural arch or of a neural arch, the centrum of which has fused to the occiput. Three myosepta, the primitive number for actinopterygians, attach to the occiput in polypterids, providing another source of evidence for lack of incorporation of vertebrae into the occiput. The occiput of the chondrostean Acipenser is comparable at an early developmental stage to that of polypterids in that three myosepta attach to it. Subsequently, in early development at least, the first neural arch is incorporated into the occiput. Additional arches continue to fuse to the expanded occiput in later development. In the ginglymodan Lepisosteus, three myosepta attach to the back of the skull in early development, demonstrating that its occiput is the same as in polypterids and Acipenser. The first two neural arches are incorporated into the occiput during further development, as evidenced by five myosepta attaching to the back of the skull. We found that Atractosteus differs from Lepisosteus in that only the first neural arch is incorporated into the occiput, evidenced by only four myosepta attaching to the back of the skull. The occiput of early developmental stages of Amiahas four myosepta attaching and a free centrumless first neural arch, showing that at least part of a vertebra has been incorporated. This hypothesis gains support from the presence of haemal processes at the posterior base of the occiput, which are otherwise present only on vertebrae. We discuss Fürbringer’s work on the spino-occipital nerves of gnathostomes, point out its shortcomings and argue that his conclusions about the great variation in number of vertebrae fused to the back of the skull in gnathostomes is a myth. We conclude that a major confounding factor in the interpretation of the occiput in actinopterygians has been the failure to distinguish between evolutionary and developmental incorporation of vertebrae.

G. David JOHNSON and Ralf BRITZ:
Occipito-vertebral fusion in actinopterygians: conjecture, myth and reality. Part 2: Teleosts
[pp. 95-110, 6 coloured figures]

The composition of the occiput of teleosts has been the subject of much conjecture for more than a century. We review various lines of putative evidence that have been presented in support of the hypothesis that one or more vertebrae (or parts thereof) have fused with the occiput in one or another teleost. As in Part 1of this paper, which deals with nonteleosts, we maintain that ontogeny and attachment of anterior myosepta provide the only unequivocal landmarks for elucidating the composition of the occiput. With this fundamental approach, we show that Heterotis and Megalops are the only teleosts, aside from molids, ostraciids and male cetomimids, in which a vertebral centrum is fused ontogenetically to the occiput. We review and clarify the distribution of the so-called accessory neural arch, ANA, long touted as evidence for occipito-vertebral fusion in teleosts and demonstrate that it has nothing to do with incorporation of the first centrum of the vertebral column. The ANA develops in the posteriormost occipital myoseptum, the third, and is never associated with a centrum, whereas the first centrum and its respective neural arch develop in the fourth myoseptum. We find no reason to question the primary homology of the ANA of Elops with that of clupeocephalans and conclude that the level at which it may be synapomorphous can only be determined by parsimony argumentation.

Lionel CAVIN:
The Late Jurassic ray-finned fish peak of diversity: biological radiation or preservational bias?
[pp. 111-121, 4 black-and-white figures, 1 table]

A diversity peak of ray-finned fishes (actinopterygians) in the Late Jurassic is revealed by the fossil record. The peak is observed for actinopterygians as a whole, as well as for marine actinopterygians only. However, diversity curves based on compilation of fossil occurrences do not directly mirror biological diversity because several biases affect the fossil record and its study. A way to distinguish between sampling and biologically driven patterns is to look at the phylogenetic relationships of the contemporaneous taxa composing the curves. If taxa constituting a peak of diversity are closely related (derived from recent cladogenetic events), they likely result from a biological radiation, while if they belong to lineages not closely related (phylogenetically distant and/or from ancient cladogenetic events), the peak is probably due to preservational bias. Distinguishing both situations can be made by measuring the average ghost lineage duration of all taxa known in a time interval. In the Late Jurassic, the average ghost lineage duration does not drop while the diversity rises, indicating that the peak of diversity of that time interval is due to a preservational bias, or Lagerstätten effect, but not associated with a biological radiation of actinopterygians. The Late Jurassic peak is compared to the mid-Cretaceous peak of diversity for marine taxa only, which is caused by a genuine biological radiation as shows the drop of the average ghost lineage duration. Recognition of the mid-Cretaceous event as a genuine biological radiation, during a period of high sea temperature, and the Late Jurassic event as a preservational bias, during a period of rather low sea temperatures, reinforces the previous observation that marine ray-finned fishes diversity is positively correlated with sea temperature.

E. O. WILEY and G. David JOHNSON:
A teleost classification based on monophyletic groups
[pp. 123-182]

One hundred and eighteen major groups of teleost fishes are recognized as monophyletic on the basis of morphological synapomorphies. One hundred and twelve are formally classified. Only three of 54 orders and a single suborder are not supported as monophyletic groups. A Linnaean classification is proposed that incorporates new ranks and suffix endings that avoid ambiguity as to name ending and associated rank. The classification is considered an initial effort to consistently reorganize teleost classification using synapomorphies to circumscribe monophyletic groups.

Peter L. FOREY and John G. MAISEY:
Structure and relationships of †Brannerion (Albuloidei), an Early Cretaceous teleost from Brazil
[pp. 183-218, 14 black-and-white figures, 2 tables]

The basal teleost †Brannerion Jordan, 1919, from the Aptian-Albian Santana Formation of Brazil, is redescribed on the basis of many acid-prepared specimens. Two forms are recognized, primarily on the basis of differences in the form of the parasphenoid and associated dentition, as well as from the shape and proportions of the basisphenoid. Neither form can be referred with confidence to the two existing nominal species because their holotypes do not show these parts of their anatomy; hence our forms are simply referred to as “sp. A” and “sp. B”. A computer cladistic analysis using 74 characters and 24 basal teleost taxa and Amia as the outgroup places †Brannerion as a basal albulid in an unresolved trichotomy with †Lebonichthys and †Baugeichthys. Uncertain relationships are due to lack of evidence, missing values and homoplasy. †Brannerion nevertheless remains one of the most completely known fossil albulids and its basal position within the group suggests that the genus can be used in wider studies of lower teleost interrelationships.

Eric J. HILTON and Ralf BRITZ:
The caudal skeleton of osteoglossomorph fishes, revisited: comparisons, homologies, and characters
[pp. 219-237, 7 coloured and 2 black-and-white figures]

The caudal skeleton is typically a rich source for phylogenetically informative characters for systematic studies of fishes, particularly among groups of basal teleosts. The caudal skeleton of Osteoglossomorpha has received a moderate amount of attention as a character system in recent phylogenetic studies of the group, as well as in studies of their systematic position among teleostean fishes as a whole. Osteoglossomorph fishes are one of the most basal lineages of teleosts with living representatives, yet most extant members are highly morphologically derived. This paper reviews the caudal skeleton of fossil and living osteoglossomorphs, and summarizes and clarifies new hypotheses of homology of its constituent elements. New information on the development of the caudal skeleton is presented for selected taxa. Among living osteoglossomorphs, the two species of Hiodon have the most plesiomorphic caudal skeletons, with seven hypurals, up to four pairs of strap-like uroneurals that extend anteriorly to pu2, and a single epural. A similar caudal skeleton is found in fossil osteoglossomorph taxa such as †Lycoptera. Most other extant osteoglossomorphs have six or fewer hypurals at any stage of ontogeny (Arapaimararely has seven). There is significant variation in the number of elements in the series (e. g., one dorsal hypural in Notopteridae and some mormyrids, to four dorsal hypurals in most osteoglossids, and five in Hiodon), as well as in fusion patterns within the hypural series (e. g., all dorsal hypurals fuse together in Arapaima and Heterotis). The Australian species of Scleropages (S. jardini and S. leichardti) are unique among osteoglossomorphs in possessing three hypurals in the ventral portion of the caudal skeleton. The “epural” of osteoglossids is actually a pair of uroneurals that fuse during ontogeny along the dorsal midline; this may subsequently fuse to the dorsal hypurals (e. g., in Arapaima and Heterotis). In most osteoglossomorphs (e. g., Hiodon, mormyrids, Arapaima, Scleropages, and Osteoglossum) the uroneurals preform in cartilage. In contrast, in Pantodon a single, unpaired element ossifies in membrane in the position of the paired uroneurals of other osteoglossomorphs. The caudal skeleton of Pantodon shares both plesiomorphic (e. g., a median uroneural element) and apomorphic (e. g., only 4 hypurals) osteoglossine features of the caudal skeleton with Osteoglossum</i>; further study of the anatomy of osteoglossines and Pantodon is needed to elucidate their relationships to each other and to other osteoglossomorph fishes.

ZHANG JIANG-YONG:
Validity of the osteoglossomorph genus †Asiatolepis and a revision of †Asiatolepis muroii(†Lycoptera muroii)
[pp. 239-249, 4 black-and-white figures]

The morphology of “†Lycoptera” muroii is reviewed based mainly on newly collected specimens from the Jingangshan Bed, Yixian Formation, of the Jehol Group in Yixian, Liaoning, China and referred to the genus †Asiatolepis as †A. muroii. Revised characters for the species are: head short; frontal short and broad; interfrontal suture sinuous; supramaxilla small; interopercle large and broad; fourth infraorbital nearly semicircular; conical teeth present on oral margin and palatine, ectopterygoid, endopterygoid, parasphenoid and basihyal; pectoral fin with I+6-7+I fin rays; pelvic fin small, with I+4 fin rays; dorsal fin relatively small, with I+7-8 fin rays; anal fin with I+9-10 fin rays; vertebrae 40-41; neural arches posterior to dorsal fin paired on some caudal centra and the two neural spines on them equal-length or with one shorter than the other; epineural fused with base of neural arch; uroneurals 4-5, hypurals 8; epural absent; caudal fin deeply forked with I+15+I principle fin rays; scales in lateral line about 36. †Asiatolepis is regarded as a valid genus and can be distinguished from †Lycoptera by at least two characters (epural absent and 15 or fewer branched caudal fin rays). †Asiatolepis is probably more primitive than †Paralycoptera and other early osteoglossomorphs in having paired neural arches posterior to the dorsal fin.

Mário DE PINNA and Fabio DI DARIO:
The branchial arches of the primitive clupeomorph fish, Denticeps clupeoides, and their phylogenetic implication (Clupeiformes, Denticipitidae)
[pp. 251-268, 9 black-and-white figures]

A detailed description is presented of the branchial-arch skeleton of the relictual clupeiform Denticeps clupeoides, recent sister group to all the remainder of the order. Information on this anatomical complex was previously available only fragmentarily, due to scarcity of study material and limitations of preparation techniques. A new look at the branchial-arch skeleton reveals several phylogenetically relevant traits at different levels, providing new characters supporting the position of Denticeps as sister group to all other clupeiforms. For example, in all clupeoids, the first through third ceratobranchials each have a single row of gill rakers, an apomorphic condition, while Denticeps retains the plesiomorphic condition of a double row. Also, the close proximity of upper branchial-arch elements typical of most clupeoids is not seen in Denticeps. Finally, the second and third infrapharyngobranchials in Denticeps are short, not apomorphically produced anteriorly as narrow long processes as in all clupeoids. A vestigial cartilage on the distal tip of the fifth ceratobranchial is newly reported in Denticeps and considered as possibly homologous with the accessory cartilage seen in some other lower teleosts but previously thought to be absent in clupeomorphs. This provides circumstantial support for a possible relationship between Otocephala and argentiniforms and may indicate a vestigial remnant of an epibranchial organ in Denticeps.

Francisco José POYATO-ARIZA, Terry GRANDE and Rui DIOGO:
General overview of fossil and Recent Gonorynchiformes (Teleostei, Ostariophysi)
[pp. 269-293, 10 black-and-white figures, 1 appendix]

The Gonorynchiformes are an intriguing group of teleosts. Most evidence indicates that they are the sister group of the Otophysi (i. e., fishes with a functioning Weberian apparatus), forming together the clade Ostariophysi. The Gonorynchiformes consist of {Chanidae + [Gonorynchidae + Kneriidae]}. Although they comprise only about 24 genera, they show amazing variety in morphology, distribution, ecology, and behaviour. There is only one Recent chanid genus (Chanos) and one Recent gonorynchid genus (Gonorynchus); all other Recent genera are kneriids. Chanids and gonorynchids present a consistent fossil record, whereas kneriids lack one. The fossil record of the Gonorynchiformes, as an ensemble, is widespread for such a small group, extending to Europe, North and South America, Africa, and Asia; it dates back to the Early Cretaceous, so that gonorynchiforms may be considered a relict or ‘living fossil’ group. We herein provide a succinct, introductory overview of this order. A historical summary on the recognition of their kinship and their phylogenetic placement among primitive teleosteans is given. We especially focus on the taxonomic diversity of the Gonorynchiformes, both in Recent and fossil forms. The morphological characterization and the evolutionary history of its three families are also outlined.

Kevin W. CONWAY, M. Vincent HIRT, Lei YANG, Richard L. MAYDEN and Andrew M. SIMONS:
Cypriniformes: systematics and paleontology
[pp. 295-316, 3 coloured and 3 black-and-white figures, 1 table]

The order Cypriniformes is a large monophyletic group of primary freshwater fishes. Nine morphological characters that support the hypothesis of monophyly of the order are reviewed and described. The order is divided into two clades, the superfamily Cyprinioidea containing the family Cyprinidae and the superfamily Cobitoidea containing seven families: Balitoridae, Botiidae, Catostomidae, Cobitidae, Gyrinocheilidae, Nemacheilidae, and Vaillantellidae. Morphological support for monophyly of all families is described with the exception of the Nemacheilidae; however molecular data supports monophyly of this group. Phylogenetic relationships among the families are largely resolved with the exception of a basal trichotomy in Cobitoidea including Catostomidae, Gyrinocheilidae, and the remaining cobitoids. Relationships within the largest family, the Cyprinidae, are unclear and the morphological and molecular studies performed to date provide conflicting hypotheses of relationship. The fossil record of cypriniform fishes dates to the Early Paleocene (62 million years ago) and is largely restricted to catostomid and cyprinid fossils.

Maria Claudia MALABARBA and Luiz R. MALABARBA:
Biogeography of Characiformes: an evaluation of the available information of fossil and extant taxa
[pp. 317-336, 7 black-and-white figures]

Characiformes have long been considered a group of relevance to understanding earth history because of their confinement to freshwater. The phylogenetic history of the group, however, is still poorly understood, preventing the formulation of a reliable biogeographical hypothesis. The accepted cladistic hypothesis of relationships to other ostariophysan orders and even the monophyly of the order have been challenged in recent molecular studies. However, most of these studies are based on very low numbers of characiform and other ostariophysan representatives, and may be affected by the long-branch-attraction artifact, as exemplified in suggestions of gonorynchiform relationships to both ostariophysans and clupeomorphs. Previous hypotheses to explain the current distribution of the Characiformes are re-evaluated. Marine origin or marine dispersal, along with massive extinction of characiform lineages in Africa, is not supported. Instead, we suggest that absence of some lineages and lower diversity overall of characiforms in Africa may be the result of vicariance together with distribution patterns of ancient characiform lineages within Gondwana.

Dominique ADRIAENS, Jonathan N. BASKIN and Hendrik COPPENS:
Evolutionary morphology of trichomycterid catfishes: about hanging on and digging in
[pp. 337-362, 8 coloured and 2 black-and-white figures]

The catfishes (Siluriformes) comprise a particularly diverse teleost clade, from a taxonomic, morphological, biogeographical, ecological and behavioural perspective. The Neotropical Trichomycteridae (the “parasitic” catfishes) are emblematic of this diversity, including fishes with some of the most specialized habits and habitats among teleosts (e. g. hematophagy, lepidophagy, miniaturization, fossorial habitats, altitudinal extremes). Relatively little information is available on general trichomycterid morphology, as most work so far has concentrated on phylogenetically informative characters, with little concern about general descriptive anatomy. In this paper we provide a synthesis of new and previously-available data in order to build a general picture of basal crown group trichomycterid morphology and of its main modifications. We focus on the evolutionary morphology in two relatively distal trichomycterid lineages, i. e. the hematophagous Vandelliinae and the miniature, substrate dwelling Glanapteryginae. New evidence is discussed in relation to the evolution of the opercular system as well as morphological modifications in miniature species exhibiting an interstitial life style.

Jacob J. D. EGGE:
Systematics of ictalurid catfishes: a review of the evidence
[pp. 363-378, 7 black-and-white figures, 2 tables]

Ictalurid catfishes represent one of the most thoroughly studied siluriform families. Several phylogenetic hypotheses have been proposed based on a variety of data types and analytical methods. While some relationships have been consistently recovered with strong support, many other relationships remain uncertain. Cranoglanidids are the likely sister-taxon to ictalurids, but the relationship of this clade to other siluriforms remains unclear. Several hypotheses of relationship among ictalurid genera have been reported using both morphological and molecular data. Relationships among <i>Ameiurus and Noturus species are the best studied; however, different topologies have been recovered using morphological and molecular data for both genera. Several questions still surround the relationships of Ictalurus species, which have yet to be studied using molecular data. Additional phylogenetic work, including comprehensive character analysis and improved taxon sampling in molecular studies, will likely help to further elucidate ictalurid relationships. Continued work in this area should provide phylogenetic hypotheses to serve as the framework for future evolutionary analyses within the group.

Mark V. H. WILSON and Robert R. G. WILLIAMS:
Salmoniform fishes: key fossils, supertree, and possible morphological synapomorphies
[pp. 379-409, 2 coloured and 9 black-and-white figures, 2 tables]

The salmoniforms and their relatives are of great interest to ichthyologists but their relationships remain controversial. Salmoniforms and osmeriforms also have an interesting but not very ancient fossil record. We summarize the results of decades of phylogenetic research by coding trees from 34 studies published since 1982 and constructing a supertree of 59 terminal taxa. The tree is used as a starting point for a discussion of selected osteological and myological synapomorphies for certain clades (e. g., Salmonidae + Esocoidei; Osmeriformes + Neoteleostei), and other morphological evidence that contradicts certain other relationships (supposed paraphyly of Umbridae and Galaxioidei). The new taxon Zoroteleostei is named to combine the Osmeriformes (galaxioids, osmeroids, and argentinoids) together with the Neoteleostei, as indicated by some molecular studies and by their possession of an open sensory canal bordered by a bony flange on the preopercular bone. Areas of research that appear to have reached a satisfying resolution include relationships within Salmoninae, including confirmation of Brachymystax and Hucho as ‘basal’ salmonine taxa, Salvelinus as sister of Parahucho plus Salmo plus Oncorhynchus, Parahucho perryi as a distinct lineage sister to Salmo and Oncorhynchus, endemic eastern Mediterranean salmonids as derivatives of Salmo trutta, and Oncorhynchus mykiss, O. clarkii, and O. masou as relatively primitive members of Oncorhynchus. Within Osmeroidei, the status of Plecoglossidae and Salangidae as successive sister groups of Osmeridae, and of Mallotus and Hypomesus as relatively primitive members of the Osmeridae are suggested. Future research should focus on some of the areas of greatest interest and uncertainty, including the mentioned possible clades in need of further study, the still-unsettled relationships within Coregoninae and among species groups of Salvelinus, and a search for fossil evidence of the early radiation of these interesting and important fishes.

Amanda BURDI and Terry GRANDE:
Morphological development of the axial skeletons of Esox lucius and Esox masquinongy (Euteleostei: Esociformes), with comparisons in developmental and mineralization rates
[pp. 411-430, 6 coloured and 3 black-and-white figures]

The developmental morphology of the axial skeleton of Esox lucius (i. e., Northern Pike and type species of the genus) and E. masquinongy (i. e., Muskellunge) was investigated. More than 1,000 specimens were examined ranging in size from about 10 mm notochordal length (NL) post-hatching juveniles, to over 80 mm standard length (SL) foraging sub-adults. Results show that regardless of individual variation, the relative sequence of bone formation and mineralization is consistent between the two species. This consistent developmental pattern enabled us to construct an ontogenetic staging scheme of eight developmental stages, each characterized by one defining criterion. The first appearance in cartilage and/or first sign of mineralization of each axial skeleton bone was plotted against time and age for each species and compared. Observed variation in bone development (e. g., number of epurals) inconsistent with the published literature is discussed.

Based on our developmental study, Esox lucius grows in size faster than E. masquinongy, but its axial skeleton develops and mineralizes more slowly. For example, at 25 mm SL, the axial skeleton of E. masquinongy is 55 % mineralized, while E. luciusis only 25 % mineralized. Esox masquinongy at this size however, is 1000 hours old, while E. lucius is only 700 hours. These results suggest that E. masquinongy has adapted a developmental strategy whereby more emphasis is put into skeletal development than into growth in size. This strategy may reflect the early foraging behavior of E. masquinongy. Unlike E. lucius, E. masquinongy absorbs its yolk sac earlier in life, and becomes an active predator just a few days after hatching. A well-mineralized axial skeleton with developed dentition would facilitate this early predacious behavior.

Matthew P. DAVIS:
Evolutionary relationships of the Aulopiformes (Euteleostei: Cyclosquamata): a molecular and total evidence approach
[pp. 431-470, 8 black-and-white figures, 3 tables, 3 appendices]

Evolutionary relationships of the Aulopiformes (Euteleostei: Cyclosquamata) are investigated from a molecular and total evidence approach that includes previous morphological datasets. Molecular and total evidence analyses recover Aulopiformes as monophyletic and sister to a monophyletic Ctenosquamata, supporting the monophyly of Eurypterygii with molecular data. Monophyly of recently considered aulopiform suborders is tested, and Chlorophthalmoidei are deemed paraphyletic. The recently described genus Paraulopus is recovered outside Chlorophthalmus based on molecular and total evidence analyses, but is not recovered as the basal member of the Synodontoidei. Giganturoidei are recovered as the sister group of an ipnopid clade, rather than the sister group to Alepisauroidei. Molecular analyses strongly support a clade consisting of the family Scopelarchidae and chlorophthalmoid taxa, but total evidence analyses recover scopelarchids as the basal lineage of Alepisauroidei. A sister-group relationship between Evermannellidae and Scopelarchidae is not supported, and the family Paralepididae is deemed paraphyletic. Systematic placement of taxa within the monophyletic and paraphyletic suborders, revised classification, and evidence supporting previously unrecognized clades are discussed.

Irma VILA, Sergio SCOTT, Natalia LAM, Patricia ITURRA and Marco A. MÉNDEZ:
Karyological and morphological analysis of divergence among species of the killifish genus Orestias (Teleostei: Cyprinodontidae) from the southern Altiplano
[pp. 471-480, 5 black-and-white figures, 4 tables, 1 appendix]

Orestias Valenciennes, 1839, a genus of killifish classified in the family Cyprinodontidae, is endemic to the Andean high plains (Altiplano) aquatic systems. Species have been classified in four species complexes: O. cuvieri, mulleri, gilsoni, and agassizii. Previous taxonomic studies on the agassizii complex have been based mainly on external morphology. The present study describes the species of the agassizii complex of the southern Altiplano using chromosomal, meristic, and morphometric characters. Species show differences in number and morphology of chromosomes. Meristic data based on examination of large numbers of juveniles and adults show that the characters evaluated were in agreement with the original systematic descriptions; however, we detected a high degree of overlap among species. Multivariate analyses showed that O. ascotanensis, O. agassizii, O. laucaensis, and O. chungarensis could be morphologically distinguished from the other species; only O. parinacotensis and O. piacotensis were not different from each other. Our results show that the chromosomal and morphological characters are informative traits in the study of the systematics of the species of the Orestias agassizii complex.

The origin and the phylogenetic interrelationships of teleosts have been controversial subjects of interest ever since Greenwood et al. (1966) presented a revision of teleost phylogeny and Patterson (1973, 1977) proposed a relationship between teleosts and Amia (or Halecomorpha) and named the group Halecostomi. Many differing views exist on the fundamental problem of teleost origins and phylogeny (e. g., Olsen 1984, Nursall 1996, Gardiner et al. 1996, Arratia 1999, Inoue et al. 2003, and many others). Different taxa (Amia, Lepisosteus, Amia + Lepisosteus, †Pycnodontiformes, †Dapedium, †Pachycormiformes, and others) have been proposed as the sister group of teleosts. Tremendous advances have occurred in our knowledge of Halecostomi and in their major component the teleosts over the past 40 years, with many new key fossils having been studied (and many extant basal teleost clades having been traced back to the Jurassic in detailed studies by Arratia 1987, 1996, 2000). In addition to new fossils, a large number of new morphological characters have been incorporated in recent phylogenetic analyses, adding to our arsenal of approaches. However, as noted by Nelson (2006), there are still many areas of disagreement in teleost phylogeny.

In recent years, molecular characters are increasingly used for assessment of actinopterygian phylogeny, sometimes conflicting with morphological data. While potentially very numerous and powerful indicators of relationships, their greatest limitation is that only a few taxa (Acipenseriformes, Amia, Lepisosteidae – all three having been proposed as the sister group of teleosts in molecular studies) can be used in the molecular search for the closest relative of the teleosts. Recently, the validity of the Halecostomi, the Halecomorphi (Amia) + teleosts, has been questioned by morphological (Grande 2005, Hurley et al. 2007) and molecular (Kikugawa et al. 2004) investigations, and the old group Holostei, comprising the Halecomorpha and the Ginglymodi (e. g., Lepisosteus), is now thought by some investigators to be monophyletic after all.

Closely interwoven with the search for the sister group of teleosts is the question of interrelationships of basal teleosts. Both morphological studies (including new fossil taxa) and molecular studies have examined this question. Studies of Patterson (1977) and Patterson & Rosen (1977) placed the Osteoglossomorpha at the base of extant teleosts, a position given to Elopomorpha by Arratia (1991, 1996, 1999). Molecular analyses tend to agree with Patterson’s 1977 view (e. g., Inoue et al. 2003) or else they place both taxa as the sister group to all other teleosts (Lê et al. 1993) or in an unresolved relationship. Differing views exist also on the boundaries of the teleosts (explored, e. g., by Patterson & Rosen 1977, de Pinna 1996, and Arratia 2001). The names Teleocephala (of de Pinna 1996 for the crown group of all extant teleosts) and Teleosteomorpha (of Arratia 2001 for Teleostei s. str. and their closest relatives) have been proposed to variously recognize stem-based fossils of the Teleostei. These taxonomic names reflect the translation into classification of trees resulting from different levels of phylogenetic analyses.

Higher up the tree in the crown-group teleosts, Clupeomorpha and Ostariophysi have been considered by many to comprise a monophyletic group after Lê et al. (1993) and, especially, Lecointre & Nelson (1996). The resulting taxon was termed the Ostarioclupeomorpha by Arratia (1996) and the Otocephala by Johnson & Patterson (1996). This suggested relationship has not been accepted by all ichthyologists, but further testing of this hypothesis will come with new morphological and molecular data. As well as there being some uncertainty here, many other questions remain on relationships of higher teleosts, including percomorphs, providing fertile ground for investigation by ichthyologists in the future.

These questions were and are at the center of the research of Gloria Arratia. To recognize her contributions to the origin and phylogeny of teleosts, the American Society of Ichthyologists and Herpetologists (ASIH) sponsored the symposium “Origin and phylogenetic interrelationships of teleosts” organized by the three editors of this volume and held at the Society’s annual meeting in St. Louis on 14 July 2007. At the same meeting, Gloria Arratia was honored with the Robert H. Gibbs, Jr. Memorial Award, 2007, for her contributions to systematic ichthyology. The present state of phylogenetic knowledge of the origin of teleosts and the interrelationships of teleost groups, key issues in fish systematics, based on both morphological (of extant and fossil taxa) and molecular data were presented. Progress employing the characters and taxa and in establishing databases (morphological and molecular) were also presented and evaluated from different perspectives by many contributors in the symposium. Most of the talks given at that meeting form the basis of the papers collected in this volume, together with three additional contributions. The editors and authors are pleased to dedicate this volume to Gloria Arratia in honor of her contributions and in the hope that its contents will assist and stimulate future research on the subjects that interest her most.

The Editors

ADRIAENS, Dominique, Evolutionary Morphology of Vertebrates, Ghent University, Gent, Belgium.
BASKIN, Jonathan N., Biological Sciences Department, California State Polytechnic University Pomona, Pomona, California, U.S.A.
BRITZ, Ralf, Department of Zoology, The Natural History Museum, Cromwell Road, London, U.K.
BROUGHTON, Richard E., Oklahoma Biological Survey and Department of Zoology, University of Oklahoma, Norman, Oklahoma, U.S.A.
BURDI, Amanda, Department of Biology, Loyola University Chicago, Chicago, Illinois, U.S.A.
CAVIN, Lionel, Department of Geology and Palaeontology, Museum of Natural History, Geneva, Switzerland.
CONWAY, Kevin W., Department of Biology, Saint Louis University, St. Louis, Missouri, U.S.A.
COPPENS, Hendrik, Evolutionary Morphology of Vertebrates, Ghent University, Gent, Belgium.
DAVIS, Matthew P., Division of Ichthyology, Natural History Museum and Biodiversity Institute, and Department of Ecology and Evolutionary Biology, The University of Kansas, Lawrence, Kansas, U.S.A.
DE PINNA, Mário, Museu de Zoologia da Universidade de São Paulo, São Paulo, São Paulo, Brazil.
DI DARIO, Fabio, Grupo de Sistemática e Biologia Evolutiva, Núcleo em Ecologia e Desenvolvimento Sócio-Ambiental (NUPEM), Universidade Federal do Rio de Janeiro, Macaé, Rio de Janeiro, Brazil.
DIOGO, Rui, Center for the Advance Study of Hominid Paleobiology, Department of Anthropology, The George Washington University, Washington DC, Washington, U.S.A.
EGGE, Jacob J. D., Department of Biology, Pacific Lutheran University, Tacoma, Washington, U.S.A.
FOREY, Peter L., Department of Palaeontology, The Natural History Museum, South Kensington, London, U.K.
GRANDE, Terry, Department of Biology, Loyola University Chicago, Chicago, Illinois, U.S.A.
HILTON, Eric J., Department of Fisheries Science, Virginia Institute of Marine Science, College of William and Mary, Gloucester Point, Virginia, U.S.A.
HIRT, M. Vincent, Bell Museum of Natural History & Graduate Program in Ecology, Evolution and Behavior, University of Minnesota, Saint Paul, Minnesota, U.S.A.
ITURRA, Patricia, ICBM, Facultad de Medicina, Universidad de Chile, Santiago, Chile.
JOHNSON, G. David, Biodiversity Research Center, University of Kansas, Lawrence, Kansas, U.S.A.; and Division of Fishes, National Museum of Natural History, Smithsonian Institution, Washington D.C., Washington, U.S.A.
LAM, Natalia, ICBM, Facultad de Medicina, Universidad de Chile, Santiago, Chile.
MAISEY, John G., Division of Paleontology, American Museum of Natural History, New York, New York, U.S.A.
MALABARBA, Luiz R., Departamento de Zoologia, Universidade Federal do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil.
MALABARBA, Maria Claudia, Museu de Ciências e Tecnologia, Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil.
MAYDEN, Richard L., Department of Biology, Saint Louis University, St. Louis, Missouri, U.S.A.
MÉNDEZ, Marco A., Laboratorio de Genética y Evolución, Departamento de Ciencias Ecológicas, Facultad de Ciencias, Universidad de Chile, Santiago, Chile.
NELSON, Joseph S., Professor Emeritus, Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada.
NURSALL, J. Ralph, Department of Biological Sciences, University of Alberta. Private: Whaletown, British Columbia, Canada.
POYATO-ARIZA, Francisco José, Unidad de Paleontología, Departamento de Biología, Universidad Autónoma de Madrid, Cantoblanco, Madrid, Spain.
SCOTT, Sergio, Laboratorio de Limnología, Facultad de Ciencias, Universidad de Chile, Santiago, Chile.
SIMONS, Andrew M., Bell Museum of Natural History & Department of Fisheries, Wildlife, and Conservation Biology, University of Minnesota, Saint Paul, Minnesota, U.S.A.
VILA, Irma, Laboratorio de Limnología, Facultad de Ciencias, Universidad de Chile, Santiago, Chile.
WILEY, E. O., Biodiversity Research Center, University of Kansas, Lawrence, Kansas, U.S.A., and Division of Fishes, National Museum of Natural History, Smithsonian Institution, Washington D.C., Washington, U.S.A.
WILLIAMS, Robert R. G., Ottawa, Ontario, Canada.
WILSON, Mark V. H., Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada.
YANG, Lei, Department of Biology, Saint Louis University, St. Louis, Missouri, U.S.A.
ZHANG Jiang-yong, Laboratory of Evolutionary Systematics of Vertebrates, Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences, Beijing, China.