Figures
Abstract
The Hapalogenys genus was the most controversial and problematic in phylogenetic position of Percoidei. In this study, we rechecked the taxonomic status of Hapalogenys in Percoidei using complete mitochondrial genome data. We purposefully added a new complete mitochondrial sequence from chosen species of Hapalogenys and conducted phylogenetic analyses using a large complete mitochondrial data set. The resultant tree topologies were congruent from partitioned Bayesian and Maximum-likelihood methods. The results indicated that Hapalogenys was distantly related to Haemulidae and could be removed from Haemulidae. The results supported the Hapalogeny was upgraded to a family rank titled Hapalogenyidae, and it should be recognized in a separate family of Hapalogenyidae. A relaxed molecular-clock Bayesian analysis indicated that the divergence times of Perciformes groups had a much older than the available old fossil records. The origin of the common ancestral lineage of Perciformes fish was estimated in the late Jurassic about 149 Myr ago.
Citation: Wei T, Sun Y, Zhang B, Wang R, Xu T (2014) A Mitogenomic Perspective on the Phylogenetic Position of the Hapalogenys Genus (Acanthopterygii: Perciformes) and the Evolutionary Origin of Perciformes. PLoS ONE 9(7): e103011. https://doi.org/10.1371/journal.pone.0103011
Editor: Valerio Ketmaier, Institute of Biochemistry and Biology, Germany
Received: April 23, 2014; Accepted: June 26, 2014; Published: July 31, 2014
Copyright: © 2014 Wei 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.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All sequence files are available from the GenBank database accession number NC019646.
Funding: This study was supported by National Natural Science Foundation of China (31272661), Zhejiang Provincial Natural Science Foundation (LY13C040001), and grant of Young Academic Leaders in Zhejiang Province (PD2013220). 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.
Introduction
The largest order of vertebrate, the Perciformes is the most diversified teleost, and it has approximate 9000 extant species placed into 18 suborders. The most species of Perciformes occur in both marine and freshwater areas ranging from shallow freshwater to depths of more than 2300 metres in the ocean [1]. Owning to some species possessing scientific and economic importance to fishery, the Perciformes fish has gained much attention in the aspect of molecular phylogenetics and systematics [2], [3]. Currently, the Perciformes order is divided into 148 families, including about 1367 genera [1]. Among all the genera of the order, the Hapalogenys genus is one of the most controversial and problematic about its phylogenetic position in Percoidei. The genus contains about 12 recognized species, such as Hapalogenys analis, Hapalogenys kishinouyei and Hapalogenys nigripinnis. It has been traditionally classified in the Haemulidae family because of the presence of chin pores in the order Perciformes [4]. All of the species typical inhabit sandy bottoms or muddy rocky shores along the tropical and temperate coast regions of West Pacific. However, some species have become very difficult to be found so far.
There has been some dispute in the literature as to genus Hapalogenys and whether it should be treated as separate families [4]–[6]. At the morphological level, the research of Hapalogenys is poorly understood, and its accurate taxomomic status remains ambiguity. Meanwhile, some evaluations of the classification and phylogeny of Hapalogenys have investigated the evolution of the complex genus. Different molecular markers, such as nuclear TMO-4c4 gene or mitochondrial genes (12s, 16s rRNA, COI and Cytb) have been used for evaluating phylogenetic relationships within the Hapalogenys. Some molecular examples of solving relationships of Hapalogenys include studies by Zhu et al. (2006), Ren et al. (2007), Ren and Zhang (2007), Liang et al. (2010), Xu et al. (2010), and Tavera et al. (2012) (Fig. 1) [6]–[11]. Each of these studies has provided valuable information for phylogenetic relationships of Hapalogenys fish. Zhu et al. (2006) proposed Hapalogenys belong to Haemulidae family, and it had a close genetic relationship with Plectorhynchus. However, later, Ren et al (2007) and Ren and Zhang’s (2007) results suggested the genus Hapalogenys was removed from the family Haemulidae according to their molecular findings. In 2010, Xu et al. advised that Hapalogenys had surpassed the genetic differentiation level with other Haemulidae species. However, it was still belong to Percoidea for the genetic distance and phylogenetic analysis. Liang et al. (2012) and Tavera et al. (2012) suggested Hapalogenys could potentially be removed from the Haemulidae family. The proposed hypotheses based on the short sequences were contradictory and controversial due to different analytical methods and datasets. The phylogeny of the Hapalogenys has never been studied by using a longer mitogenomic dataset and a robust phylogenetic tree with appropriate representation species. Recently, it has been demonstrated that the complete mitochondrial genomes can offer sufficient resolution for constructing a robust phylogeney and estimating divergence time events compared with single gene sequences or segments [12], [13]. Therefore, to make a further toward the comprehension for Hapalogenys phylogeny, it was appropriate to address these questions from a longer mitogenomic perspective.
(A) Parsimony and Neighbor-Joining tree for partial mtDNA cytb gene (Zhu et al., 2006), (B) Minimum-Evolution tree for mtDNA 16s RNA gene (Ren et al., 2007), (C) Neighbor-Joining tree for mtDNA 16s RNA gene (Ren and Zhang, 2007), (D) Neighbor-Joining tree for mtDNA cytb gene (Xu et al., 2010), (E) Bayesian, Maximux-likehood, and Parsimony tree for mtDNA 16s RNA gene (Liang et al., 2012), (F) Bayesian, Maximux-likehood, and Parsimony tree for nuclear TMO-4c4 gene (Liang et al., 2012), (G) Bayesian, Maximux-likehood, and Parsimony tree for mtDNA 16s RNA and nuclear TMO-4c4 gene (Liang et al., 2012), (H) Bayesian and Maximux-likehood tree for mtDNA 16s, COI, and Cytb gene (Tavera et al., 2012).
The evolutionary history of the Perciformes fishes has received little attention oweing to scarce representation in the fossil record. Although some percoid family’s fossil was record [14], the evolutionary history and systematics of the Perciformes remained obscured. The recent technical developments in the molecular estimation of divergence times have provided calculable time scales for molecular phylogenetic trees. It can present a new approach to elaborate evolutionary history which can not be estimated by the fossil record alone [15], [16]. Therefore, we make the effort to generate new viewpoints to get insight into the origin time of Perciformes from time information of molecular data.
In this study¸ in order to address questions about the phylogenetic position of the Hapalogenys genus and the evolutionary origin of Perciformes, we obtained the complete mitochondrial genome sequence of a species of Hapalogenys for the first time. The unambiguously aligned mitochondrial genomes were used for implementing phylogenetic analysis to illustrate the phylogenetic position of the Hapalogenys genus in Percoidei. In addition, we estimated the divergence time of Perciformes using a relaxed molecular-clock method to clarify the time scale of evolution of the group.
Materials and Methods
Sampling of specimens and mitogenome sequencing
Hapalogenys analis was purchased at a local fish market (Zhoushan, China). Taxonomic status of the fishes was identified by morphology. The muscle tissue was excised and stored in 95% ethanol and frozen at −20°C. We determined the complete mitogenome sequences of H. analis and then combined with all previously published mitogenome sequences of Percoidei taxa (http://mitofish.aori.u-tokyo.ac.jp/).
Molecular Methods
Total genomic DNA was extracted from the approximated 20 mg of tissues using the Tiangen DNA extraction kit according to the manufacturer’s protocol. PCR amplification and sequencing were performed as described in Wei et al. [17]. The sequencing reactions for each of the PCR product were implemented in both directions, and the sequences of overlapping fragments were checked and assembled to determine the complete mitochondrial genome sequence. The complete mitochondrial genome was sequenced using primers in Table S1. In addition, the taxa lists in this study were shown in Table S2 coupled with GenBank accession numbers.
Sequence editing and alignment
The mitochondrial genome sequences obtained were edited and analyzed by using DNAstar and DNASIS 3.2 (Hitachi Software Engineering). We combined the newly determined sequences with 61 previously published mitogenomes to construct data sets. All sequences from the L-strand-encoded genes were switched to complementary strand sequences. Amino acid sequences were used for aligning protein-coding genes. After alignment, they were translated back to nucleotide sequences [18]. All positions comprised of gaps and stop codons were not including from the subsequent phylogenetic analysis. The ND6 gene was not applied to the phylogenetic analysis owing to its poor phylogenetic performance and heterogeneous base composition [19], [20]. The alignment of tRNA genes were performed according to the secondary structural model [18]. The 12S and 16S rRNA genes were firstly aligned using Mega 5.0 with default parameters [21] and then inspected manually using MacClade ver 4.08 for any misalignments [22]. In addition, the control region was excluded because positional homology was not determined in distantly related species. The final data set was comprised of 10476 positions from the first, second and third codon positions of the 12 protein-coding genes, 2349 positions from the two rRNA genes, and 1410 positions from the tRNA genes for phylogenetic analysis.
Phylogenetic analysis
The phylogenetic trees were performed by using partitioned Maximum likelihood and Bayesian inference. The most suitable model of DNA substitution was chosen with the jModeltest program by the Akaike Information Criterion (AIC) [23]. The partitioned Bayesian phylogenetic analyses were conducted by using MrBayes 3.2 [24]. The Markov chain Monte Carlo (MCMC) analyses (with random starting trees) was run for 3 million generations with one cold and three heated chains. We performed two independent runs for 3 million generations, with tree sampling every 100 generations. Runs were stopped after the standard deviation of split frequencies fell below 0.01. In addition, we employed the above datasets to the partitioned Maximum-likelihood (ML) analysis using RAxMLHPC [25]. For the dataset, exploring of the best scoring ML tree and a rapid bootstrap analysis were investigated with a general time reversible nucleotide model (GTR) with sites following a discrete gamma distribution (Γ) and some sities invariable (I). Nodal support of the tree was performed by 1000 replications.
Estimation of divergence times
A relaxed molecular clock Bayesian method used in MCMCTREE program of PAML was implemented for dating analysis [26]. All time constrains were used with a unit of 100 Ma (1 = 100 Ma), because some model components were scale-variant in the Bayesian analysis. Including multiple fossil-based calibration points (Table S3) were employed for diverse teleostean lineages [13], [15], [27], [28]. The independent-rates (IR) model was used for assigning the prior of rates within internal nodes. It has been thought that it is more appropriate than the autocorrelated-rates model in divergence time estimation [29]. In order to judge possible failure of the Markov chains to attain stationarity, at least two Markov chain Monte Carlo (MCMC) analyses were utilized with two different seeds for each analysis. Each MCMC analysis approximately with a burn-in period of 10,000 cycles was performed, and every 100 cycles was used for generating a total of 10,000 samples. The similar results were obtained and observed from the two runs.
Results and Discussion
Phylogenetic position of Hapalogenys
The present study was based on the complete mitochondrial genome sequences from some chosen species which represented the diversity of the Perciformes species. Partitioned Maximum-likelihood and Bayesian analysis based on complete mitochondrial genome yielded essentially the same tree topology (Fig. 2). The resultant trees indicated that Hapalogenys was found to be monophyletic and had been most closely relationship with Lethrinidae. The Hapalogenys displayed a high degree of molecular divergence, which was similar to detected morphological difference. It was different from groups of the Haemulidae herein. It had been removed from Haemulidae family, and formed a distinct group outside the Haemulidae group. The Haemulidae was resolved as a sister group relative to Mondactylidae. It had a distant phylogenetic relationship with Hapalogenys genus.
Numbers near internal branches indicate the ML bootstrap support values (left) and Bayesian posterior probabilities (right), respectively. Support values less than 50% for the node are indicated by a dash.
The results of these analyses were different from the results of previous phylogenetic studies. Traditionally, the genus Hapalogenys was placed in the Haemulidae family among Perciformes owning to the presence of pores [30]. But Johnson (1984) took the genus as “incertae sedis” in the Percoidei due to its uncertain affinities [31]. Springer and Raasch proposed a new family name for the Hapalogenys in 1995, but there was not any strong supporting evidence for the genus [5]. McKay (2001) retained Hapalogenys genus in Haemulidae for convenience. In fact, he had also realized the genus should be removed from the Haemulidae [32]. However, the phylogenetic relationship of Pomadasyidae fish was analyzed using the cytochrome b genes molecular marker by Zhu et al. (2006) [6]. He proposed that Hapalogenys and Plectorhynchus had a close genetic relationship with Pomadasys. Significantly, the 16S rRNA gene datasets were used to infer phylogenetic relationships among different genera of Haemulidae, the results suggested the genus Hapalogenys was removed from the family Haemulidae and should be redefined [7], [8]. The most recent phylogenetic analysis of Hapalogenys also provided similar result by using partial mitochondrial and nuclear genes [10], [11]. However, deep nodes were weak in constructing phylogenetic trees. It indicated that the short sequence had difficult in resolving the particular phylogenetic problem.
The most recent morphological research of the phylogenetic position of Hapalogenys was derived from having the combination of characteristics [9]. The results indicated that Hapalogenys genus differed from the Haemulidae group at morphological level. But previous morphological studies had put forward the contrary, that Hapalogenys was more likely affiliated with Haemulidae family among Perciforms [1], [33]. The phylogenetic analyses of plentiful mitochondrial genome sequence data presented herein, and it suggested the genus Hapalogenys should be removed from Haemulidae. Our molecular data was inconsistent with traditional morphological data, but was agreed with some previous molecular hypotheses. According to these results, we supported that the Hapalogenys genus was upgraded to a family rank titled Hapalogenyidae and should be recognized in a separate family of Hapalogenyidae.
Divergence Time Estimation
In present study, overall MCMCTREE analyses of the divergence time were based on the assumption of independent rates (IR). The relaxed molecular-clock Bayesian analysis of divergence time estimates suggested that the Perciformes fish was estimated to be diverged from the ancestral lineage of the Gasterosteiformes and Scorpaeniformes during the about 149 Ma with a credible interval of 133–166 Ma (Fig. 3). The common ancestor of the order has diverged into the different family on 145 Ma. Previous oldest Perciformes fossil was discovered from the early Eocene. Paleontological records indicated that Perciformes fish originated in the late Cretaceous about 90 Mya. Some families among Perciformes fastly radiated in the Paleocene or early Eocene after the Cretaceous/Tertiary boundary [34]. Benton (1993) analyzed the radiation of Perciformes fishes based on the oldest fossil records. The results showed common ancestor of the Perciformes fish dated back to over 149 Mya in the Jurassic. Meanwhile, according to Benton’s (1998) fossil records, it suggested that the radiation of Perciformes species traced back to the early Cenozoic [35], but the molecular clock estimation time take it out from the Jurassic to the early Cretaceous by Kumazawa et al. (1999) [36]. Cantatore et al. (1994) estimated divergence times of Perciformes fish based on mitochondrial Cytb genes by using molecular-clock approach. The acquired divergence times were similar to those in this study [37]. There seemed to be a significant time gap between divergence estimates from molecular evidence and the first occurrence evidence of fossil records. One explanation might be provided for apparent discrepancies between fossil chronologies and molecular data. In Perciformes families, some fossil records of fish were completely absence, and a number of the first occurrences with recognizable fossil records were based merely on the otoliths [36], [38]. Meanwhile, owning to the less preservable nature of fish fossils, disappearance or occurrence of fish was inclined to be affected by limited fossil localities of unusual preservation [35]. These questions might lead to underestimate origination of fish, and cause prejudice in deducing evolutionary history of Perciformes fishes. So far, some studies about the molecular divergence time estimates exceeded fossil record have also been found in vertebrates [16], [39], [40]. Therefore, we considered that the lacking of teleostean fossil record might lead to the discrepancy between fossil and molecular dates, rather than the inaccuracy of our molecular time estimates. Of particular interest for divergence time studies, the common ancestral lineage of Perciformes fish originated in the late Jurassic about 149 Ma (Fig. 3), and it would undergo the Cretaceous mass extinction events. Major climatic and ecosystems changes occurred during Cretaceous time resulted in extinction of most archaic actinopterygian groups. The survival of the ancestor of Perciformes through the extinction events resulted in the subsequent familial radiation during the Cretaceous-Tertiary. These time estimates were roughly in accordance with those of the previous molecular study [36]. Our results proposed that a large wave of familial radiation for Perciformes occurred during the Cretaceous-Tertiary. More studies are required to resolve the relationships between Perciformes fish radiations and global changes at the Cretaceous-Tertiary.
Divergence times were estimated from the partitioned Bayesian analysis using PAML program package. The horizontal bars represent the estimated 95% credibility intervals of the divergence time estimation.
Supporting Information
Table S2.
List of the species used in this study with DDBJ/EMBL/GenBank accession numbers.
https://doi.org/10.1371/journal.pone.0103011.s002
(DOC)
Table S3.
List of time constraints used in divergence time estimation.
https://doi.org/10.1371/journal.pone.0103011.s003
(DOC)
Author Contributions
Conceived and designed the experiments: TW TX. Performed the experiments: TW YS BZ. Analyzed the data: TW TX. Contributed reagents/materials/analysis tools: RW YS. Contributed to the writing of the manuscript: TW TX.
References
- 1.
Nelson JS (2006) Fishes of the World, fourth ed. Hoboken-New Jersey: John Willey and Sons.
- 2. Miller TL, Cribb TH (2007) Phylogenetic relationships of some common Indo-Pacific snappers (Perciformes: Lutjanidae) based on mitochondrial DNA sequences, with comments on the taxonomic position of the Caesioninae. Mol Phylogenet Evol 44: 450–460.
- 3. Song CB, Near TJ, Page LM (1998) Phylogenetic Relations among Percid Fishes as inferred from Mitochondrial Cytochrome b DNA Sequence Data. Mol Phylogenet Evol 10: 343–353.
- 4. Iwatsuki Y, Russell BC (2006) Revision of the genus Hapalogenys (Teleostei: Perciformes) with two new species from the Indo-West Pacific. Memoirs of Museum Victoria 63: 29–46.
- 5. Springer VG, Raasch MS (1995) Fishes, angling, and finfish fisheries on stamps of the world. American Topical Association, Handbook 129: 1–110.
- 6. Zhu SH, Zheng WJ, Zou JX, Yang YC, Shen XQ, et al. (2006) Molecular phylogenetic analysis of five Pomadasyidae fish based on mitochondrial cytohrome b sequences. J Trop Oceanogr 4: 42–45.
- 7. Ren G, Zhang Q, Qian KC, Xu ZN, Liu XT (2007) sequence analysis of twelve grunt fishes based on 16s ribosomal RNA gene fragments. Acta Zoot Sin 26: 48–52.
- 8. Ren G, Zhang Q (2007) Phylogeny of haemulid with discussion on systematic position of the genus Hapalogenys. Acta Zoot Sin 32: 835–841.
- 9. Liang RS, Zhou XL, Yang GH, Luo DJ, Zhong S, et al. (2012) Molecular phylogenetic relationships of family Haemulidae (Perciformes: Percoidei) and the related species based on mitochondrial and nuclear genes. Mitochondrial DNA 4: 264–277.
- 10. Xu TJ, Wang JX, Sun YN, Shi G, Wang RX (2010) Phylogeny of Hapalogenys with discussion on its systematic position in Percoidea using cytochrome b gene sequences. Acta Zoot Sin 35: 530–536.
- 11. Tavera JJ, Acero A, Balart AA, Bernardi G (2012) Molecular phylogeny of grunts (Teleostei, Haemulidae), with an emphasis on the ecology, evolution, and speciation history of New World species. BMC Evol Biol 12: 57.
- 12. Lavoué S, Miya M, Nishida M (2010) Mitochondrial phylogenomics of anchovies (family Engraulidae) and recurrent origins of pronounced miniaturization in the order Clupeiformes. Mol Phylogenet Evol 56: 480–485.
- 13. Nakatani M, Miya M, Mabuchi K, Saitoh K, Nishida M (2011) Evolutionary history of Otophysi (Teleostei), a major clade of the modern freshwaterfishes: Pangaean origin and Mesozoic radiation. BMC Evol Biol 11: 177.
- 14. Otero O (2004) Anatomy, systematics and phylogeny of both Recent and fossil latid fishes (Teleostei, Perciformes, Latidae). Zoolo J Soc-lond 141: 81–133.
- 15. Azuma Y, Kumazawa Y, Miya M, Mabuchi K, Nishida M (2008) Mitogenomic evaluation of the historical biogeography of cichlids toward reliable dating of teleostean divergences. BMC Evol Biol 8: 215.
- 16. Inoue JG, Kumazawa Y, Miya M, Nishida M (2009) The historical biogeography of the freshwater knifefishes using mitogenomic approaches: a Mesozoic origin of the Asian notopterids (Actinopterygii: Osteoglossomorpha). Mol Phylogenet Evol 51: 486–499.
- 17. Wei T, Jin XX, Xu TJ (2013) The first complete mitochondrial genome from Bostrychus genus (Bostrychus sinensis) and partitioned Bayesian analysis of Eleotridae. J Genet 92: 247–257.
- 18. Kumazawa Y, Nishida M (1993) Sequence evolution of mitochondrial tRNA genes and deep-branch animal phylogenetics. J Mol Evol 37: 380–398.
- 19. Miya M, Nishida M (2000) Use of mitogenomic information in teleostean molecular phylogenetics: a tree-based exploration under the maximum-parsimony optimality criterion. Mol Phylogenet Evol 17: 437–455.
- 20. Yoder AD, Yang Z (2000) Estimation of primate speciation dates using local molecular clocks. Mol Biol Evol 17: 1081–1090.
- 21. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Phylogenet and Evol 28: 2731–2739.
- 22.
Maddison WP, Maddison DR (2000) MacClade 4.0: analysis of phylogeny and character evolution. Sunderland: Sinauer Associates.
- 23. Posada D (2008) jModelTest: Phylogeneticmodel averaging. Mol Biol Evol 25: 1253–1256.
- 24. Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755.
- 25. Stamatakis A (2006) RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22: 2688–2690.
- 26. Yang Z (2007) PAML 4: phylogenetic analysis by maximum likelihood. Mol Biol Evol 24: 1586–1591.
- 27.
Benton MJ (1993) The Fossil record 2. Chapman and Hall, London.
- 28. 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: 58.
- 29. Zhong B, Yonezawa T, Zhong Y, Hasegawa M (2009) Episodic evolution and adaptation of chloroplast genomes in ancestral grasses. PLoS ONE 4(4): e5297.
- 30. Richardson J (1844) Description of a genus of Chinese fish. J Nat Hist 13: 462–464.
- 31.
Johnson GD (1984) Percoidei: Development and relationships. In: Moser HG, Richards WJ, Cohen DM, Fahay MP, Kendall AWJr, Richardson SL, editors. Ontogeny and systematics of fishes. Am Soc Ichthyol Herpetol Spec Publ 1.
- 32.
McKay RJ (2001) Family Haemulidae in: Carpenter K, Niem VH, (eds.) Species identification guide for fishery purposes. The living marine resources of the western central Pacific. Bony fishes part 3 (Menidae to Pomacentridae). Rome: FAO.
- 33.
Randall JE, Lim KKP (2000) A checklist of the fishes of the South China Sea. Raffles Bull Zool (Suppl. 8): 569–667.
- 34.
Carroll RL (1997) Patterns and processes of vertebrate evolution. Cambridge Univ Press, New York.
- 35.
Benton MJ (1998) The quality of the fossil record of the vertebrates. In: Donovan SK, Paul CRC (Eds) The adequacy of the fossil record. John Wiley and Sons, New York, 269–303.
- 36.
Kumazawa Y, Yamaguchi M, Nishida M (1999) Mitochondrial Molecular Clocks and the Origin of Euteleostean Biodiversity: Familial Radiation of Perciforms May Have Predated the Cretaceous/Tertiary Boundary. In: The biology of biodiversity. Springer Japan 35–52.
- 37. Cantatore P, Roberti M, Pesole G, Ludovico A, Milella F, et al. (1994) Evolutionary Analysis of Cytochrome b Sequences in Some Perciformes: Evidence for a Slower Rate of Evolution Than in Mammals. J Mol Evol 39: 589–597.
- 38. Patterson C (1993) An overview of the early fossil record of acanthomorphs. B Mar Sci 52: 29–59.
- 39. Yamanoue Y, Miya M, Inoue JG, Matsuura K, Nishida M (2006) The mitochondrial genome of spotted green pufferfish Tetraodon nigroviridis (Teleostei: Tetraodontiformes) and divergence time estimation among model organisms in fishes. Genes Genet Sys 81: 29–39.
- 40. Hurley IA, Mueller RL, Dunn KA, Schmidt EJ, Friedman M, et al. (2007) A new time-scale for ray-finned fish evolution. P Roy Soc B–Biol Sci 274: 489–498.