Chromodorid nudibranchs (16 genera, 300+ species) are beautiful, brightly colored sea slugs found primarily in tropical coral reef habitats and subtropical coastal waters. The chromodorids are the most speciose family of opisthobranchs and one of the most diverse heterobranch clades. Chromodorids have the potential to be a model group with which to study diversification, color pattern evolution, are important source organisms in natural products chemistry and represent a stunning and widely compelling example of marine biodiversity. Here, we present the most complete molecular phylogeny of the chromodorid nudibranchs to date, with a broad sample of 244 specimens (142 new), representing 157 (106 new) chromodorid species, four actinocylcid species and four additional dorid species utilizing two mitochondrial markers (16s and COI). We confirmed the monophyly of the Chromodorididae and its sister group relationship with the Actinocyclidae. We were also able to, for the first time, test generic monophyly by including more than one member of all 14 of the non-monotypic chromodorid genera. Every one of these 14 traditional chromodorid genera are either non-monophyletic, or render another genus paraphyletic. Additionally, both the monotypic genera Verconia and Diversidoris are nested within clades. Based on data shown here, there are three individual species and five clades limited to the eastern Pacific and Atlantic Oceans (or just one of these ocean regions), while the majority of chromodorid clades and species are strictly Indo-Pacific in distribution. We present a new classification of the chromodorid nudibranchs. We use molecular data to untangle evolutionary relationships and retain a historical connection to traditional systematics by using generic names attached to type species as clade names.
Citation: Johnson RF, Gosliner TM (2012) Traditional Taxonomic Groupings Mask Evolutionary History: A Molecular Phylogeny and New Classification of the Chromodorid Nudibranchs. PLoS ONE 7(4): e33479. https://doi.org/10.1371/journal.pone.0033479
Editor: Jonathan H. Badger, J. Craig Venter Institute, United States of America
Received: August 10, 2011; Accepted: February 14, 2012; Published: April 10, 2012
Copyright: © 2012 Johnson, Gosliner. 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 work was supported by the California Academy of Sciences, the University of California Santa Cruz-Department of Ecology and Evolutionary Biology, the National Science Foundation DEB 9978155 and DEB 0329054 to TG, and an Encyclopedia of Life Rubenstein Fellowship to RJ. 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 chromodorid nudibranchs are a brightly colored, morphologically diverse and species-rich group of sea slugs. Edmunds'  stated that, “Chromodorid nudibranchs are among the most gorgeously colored of all animals”. The over 300 described species are primarily found in tropical and subtropical waters, as members of coral reef communities, specifically associated with their sponge prey. The chromodorids are the most speciose family of opisthobranchs; their numbers rival the most diverse gastropod clades, e.g Cypraeidae (∼200 spp.), Conidae (∼500 spp.), Muricidae (∼1600 spp.) and Turridae (∼4,000 spp.). The beauty and diversity of the chromodorid nudibranchs has attracted attention from scientists, divers and underwater photographers. Despite this growing interest and subsequent increased exploration into their ecology –, natural products chemistry –, color pattern evolution – and natural history –, there is not a comprehensive, well-supported phylogeny of the chromodorid nudibranchs. Species misidentifications in ecological and chemical studies can lead to incorrect conclusions, especially when one species name represents more than one lineage , –. In phylogenetic studies, genera are often represented by a limited number of species, in many cases the only the type species , but little attention has been paid to the risk of drawing incorrect conclusions if the generic groupings that serve as proxies of relationship are not monophyletic , . More comprehensive understanding of ecological, biogoegraphical and evolutionary patterns in the chromodorids is hindered by the lack of a detailed molecular phylogeny of this group and continued use of known non-monotypic names to convey relationship and information.
Historically, the classification of the Chromodorididae has been based on morphological similarity, primarily radular morphology, and has included species and genera thought to be closely related to Chromodoris, Hypselodoris and Cadlina . But there has been substantial debate over the inclusion of Cadlina in the Chromodorididae. The majority of the molecular evidence and re-evaluated morphological data suggest Cadlina should not be a considered a member of the Chromodorididae, but instead the Cadlinidae, and it will not be included here , . Most previous phylogenetic studies that have focused solely on chromodorids have used only morphological data to understand species level relationships , , –. Exceptions include , , , who used molecular data. Additionally, most phylogenetic hypotheses of relationships in the chromodorid nudibranchs either focused on only one genus, or used genera, or one representative of a genus as terminal taxa. Rudman  and Rudman and Berquist  used composite representatives for each genus when building their phylogeny. Gosliner & Johnson  used published data on the type species of each chromodorid genus in their preliminary phylogeny of the family. They did not include any other species to test the monophyly of any of the genera. Valdés , in a morphological phylogeny of all dorid nudibranchs found the chromodorids, represented by ‘Cadlina’ and ‘Chromodoris’ to be monophyletic and sister to the actinocyclids. Turner and Wilson  presented the first molecular phylogeny of the chromodorid nudibranchs. Their study included fifty-six chromodorid species, the majority of which were from the eastern Atlantic (EA) and eastern and southern Australia, and for the first time, at least one species of each of the chromodorid genera. See , , [28 32]– for all details on all specimen data used. They were the first to explicitly test the monophyly of most chromodorid genera. Although, they were the first to find evidence for non-monophyly in Chromodoris, Hypselodoris and Mexichromis , they found little support for major clades and did not propose any changes to current classifications. Johnson  confirmed these findings and with the addition of sequences from fifty-five additional chromodorid and cadlinid specimens (41 species, 26 species of which had not been included in previous molecular studies) and more dorid taxa, also showed Glossodoris and Noumea to be polyphyletic  (for a review of previous hypotheses see Figure 1). We expand on this preliminary research and include more than one species from each genus (except monotypic genera), and include for the first time, the type species of every chromodorid genus. The Indo-Pacific (IP) is home to the greatest diversity of chromodorid nudibranchs – and yet the majority of taxa from this region has not been included in any molecular studies of the group, until now (Figure 2). One of the main objectives of this work is to advance chromodorid systematics and to provide a phylogenetic framework with which our traditional use of morphological data can be examined.
A. ‘Phylogenetic scenario’ for the chromodorid genera modified from , . B. Morphological phylogeny of generic representatives for the Chromodorididae . C. Combined 16s and COI phylogram of the Chromodorididae from . D. Combined 16s and COI phylogram of the Chromodorididae and Cadlinidae from . Rudman's ‘Chromodoris group’ in red, ‘Hypselodoris group’ in blue, Cadlinella in yellow, Diversidoris (not included in , , ), Cadlina in grey and other dorids in black.
New collections (from this contribution and  in blue. GenBank specimens in red. Size of circle represents number of specimens collected in each region. Specimen details in Supplementary Table S1).
The goals of this contribution are:(1) generate a phylogeny that tests the species level relationships of the chromodorid nudibranchs and confirms the monophyly of the Chromodorididae, (2) assess the phylogenetic validity of the chromodorid genera, and (3) propose a new classification for the chromodorid nudibranchs that reflects their relationships.
Materials and Methods
Selection of Taxa
In this study and a companion study , thanks to targeted collecting trips, dedicated collectors and DNA extracted from museum collections, we were able to include specimens from throughout the Indo-Pacific (IP), the eastern Pacific (EP) and West Atlantic (WA) (Figure 2 and Table S1). We use the term Indo-Pacific to define the biogeographic region including the tropical and subtropical regions of the Indian Ocean (from the Red Sea to the east coast of South Africa) and both the western and central Pacific, but not the tropical eastern Pacific . Museum collections are an invaluable resource for biodiversity studies . We have found existing natural history collections can reduce the need for additional collecting. Our study, combined with data from  and GenBank, is unique in its wide taxonomic and geographic sampling. Because we have included both the type species of every genus and additional species of all 14 of the non-monotypic genera, we can test the monophyly every genus in the family (Table S1).
We directly sequenced 142 specimens representing 106 species. We combined these new data with all available sequences on GenBank (Table S1). Specimens and data from Johnson , GenBank accession numbers beginning with EU, are included with new data for Figure 2, but are not treated as new in the numbers of specimens sequenced for this study. In total, we analyzed data from 244 chromodorid specimens, four actinocyclid species and four additional dorid nudibranch species for a total of 165 species and 252 individual specimens. We used Doris kerguelensis as the outgroup based on preliminary analyses . The chromodorid species include at least one species from all of the genera currently classified in the family Chromodorididae. The number of species included in this analysis compared to the number of described species per genus is as follows: Ardeadoris (2/2), Cadlinella (2/3), Ceratosoma (9/13, two undescribed), Chromodoris (50/88, two undescribed), Digidentis (3/4), Diversidoris (1/1), Durvilledoris (3/4), Glossodoris (17/30, two undescribed), Hypselodoris (30/59, two undescribed), Mexichromis (7/12), Noumea (12/22), Pectenodoris (2/2), Risbecia (3/5), Thorunna (8/12), Tyrinna (2/2) and Verconia (1/1) (S1). All sequences taken from GenBank are listed with GB following the species name. We also included COI sequence from two specimens from the Moorea BioCode project in our analyses (http://bscit.berkeley.edu/biocode/). We have examined all of the new specimens included here and they are deposited in natural history museums, as indicated by catalog numbers. We never combined sequences from different individuals into chimeras representing one species; specimens included in these analyses are treated as individuals.
The majority of the specimens used in this study are part of the California Academy of Sciences Invertebrate Zoology (CASIZ) collection. We had the permission of CASIZ to take tissue samples from specimens for DNA analysis. As stated in the CASIZ collections policy: ‘No specimens will be accessioned without adequate labeling, collection notes, field notes, or other locality information, nor without appropriate legal documentation (collecting permits, export permits from country of origin, etc.) when applicable.’ We also included DNA extracted for five specimens currently deposited in the Museum National d'Histoire Naturelle (Paris Museum) and the Western Australian Museum. These tissues samples were collected during joint field trips under the agreement that the tissue could be sequenced at the California Academy of Sciences, while the specimens would remain at the respective museum. All other data used is from GenBank or the Moorea BioCode Database.
Preservation, Extraction and Amplification
Most of our samples were collected especially for molecular work and were preserved accordingly, either in 95% ETOH, SED buffer (saturated NaCl solution with EDTA and DMSO) or frozen. In addition to the specimens collected specifically for molecular study, we were also able to use museum material that was, either preserved in 70–75% EtOH or the original fixation method is unknown.
DNA extraction and PCR amplification
We initially used standard phenol-cholorform extractions ,  to extract genomic DNA and also used the Dneasy spin column extraction method (Qiagen) to extract genomic DNA from the majority of our samples. We used universal primers to amplify, using PCR, double-stranded products from both the cytochrome oxidase 1 (COI) and 16S mitochondrial genes. We targeted a 658 bp fragment of COI using Folmer et al's  universal primers and for 16s sequences, We used Palumbi's  16Sar and 16Sbr primers. We carried out the polymerase chain reaction in 25 µL reactions with one µL of genomic DNA template. We used the second (200 µL) elution from my extractions in Dneasy AE buffer as the DNA template in most reactions. If the amplification was difficult we used one µL of the first elution. For the phenol-cloroform and chelex extractions, we used dilutions of 1∶25 or 1∶50. No matter the extraction method used, we included 2.5 µL of 10× PCR buffer, 0.5 µL dNTPs (10 mM stock), 0.25 µL of each primer (25 uM stock), 0.75–0.85 µL MgCl (50 mM stock), 0.25 µL Taq (5 units/µL )-Apex, Biolase,USB HotStart- and 19.5 µL of ddH2O in each reaction tube. We ran all of the reactions on a BioRad MyCycler™ Thermocycler (software version 1.065, Bio-Rad Laboratories). COI segments were amplified with the following parameters: an initial denaturation at 94°C for three minutes, then, 39 cycles of denaturation at 94°C for 30 seconds; annealing at 46°C for 30 seconds; extension at 72°C for 60 seconds, these cycles were followed by extension at 72°C for five minutes. Partial 16s sequences were amplified with the following parameters: an initial denaturation at 94°C for three minutes, then 39 cycles of denaturation at 94°C for 30 seconds; annealing at 50–52°C for 30 seconds; extension at 72°C for 60 seconds, these cycles were followed by extension at 72°C for five minutes and 25°C for 60 seconds. We used electrophoresis to view PCR products on 0.8% TBE or TAE agarose gel stained with ethidium bromide. We cleaned successful PCR products with ExoSap-It (USB Scientific) following each product's standard protocol.
The cleaned, PCR products were copied and labeled with fluorescently dye-terminators (Big Dye 3.1 ABI) in 10 µL reactions. Each reaction contained 0.5–2 µL of cleaned PCR product, 1.63 µL of 5× reaction buffer, .5 µL of primer (10 mM stock), 0.5 µL–0.75 µL of Big Dye and water to 10 µL . These reactions were run on a Perkin Elmer 9600-GeneAmp PCR System or a BioRad MyCycler™ Thermocycler (software version 1.065, Bio-Rad Laboratories). The resulting labeled, single stranded DNA was precipitated by addition of 2.5 µL of EDTA and sequential washing and pelleting in (centrifuge details) with 100% and then 70% EtOH. The pelleted DNA was denautured for two minutes at 94°C in 13–15 µL of HiDi formamide (Applied BioSystems). The denatured, labeled DNA fragments were sequenced in both directions on the ABI 3100 and 3130 Genetic Analyzer in the Center for Comparative Genomics (formerly the Osher Laboratory for Molecular Systematics) at the California Academy of Sciences.
Sequence editing and alignment
We assembled, edited and removed primer strands from forward and reverse strands for each gene fragment sequenced using Sequencher (ver. 4.7. GeneCodes Corporation) and Genious 3.0-5.3.3 (Biomatters). We aligned the COI sequences by eye and translated the base pair data into amino acids to using MacClade 4.08  to confirm alignment accuracy. We aligned 16s sequences with MUSCLE . We then further optimized the alignments by eye using both MacClade  and Genious 3.0-5.3.3 (Biomatters).
We tested for saturation or multiple substitutions at the same site by plotting the absolute number of transitions and transversions at each codon position (1st, 2nd, 3rd) for COI and at each base pair for 16s against both uncorrected p distance and log det using PAUP  and Excel (Plots not shown).
Data sets and Phylogenetic Reconstruction
Sequence data for both genes was not obtained for every specimen we studied. We worked with two main data sets, because we wanted to test the effect of missing data on the resulting phylogeny: The two data sets were: 1) Combined 16s and COI for specimens with sequence data for both genes, 2) All 16s and COI data for all specimens (Table S1.) Both of these data sets were analyzed both including and excluding variable characters in the 16s alignment. For all of these analyses we used Doris kerguelensis as the outgroup.
We determined the best-fit model of evolution for each codon position for COI (1st, 2nd 3rd) and the 16s fragment using the AIC selection from Mr. ModelTest ver.2  for each dataset. We ran Bayesian phylogenetic analyses using Mr. Bayes 3.1.2 –. We ran a Monte Carlo-Metropolis simulation for 50,000,000 generations for each dataset, with trees sampled every 1000 generations. Data was partitioned by gene and by codon position in all combined analyses. We ran one analysis of two runs of six chains for all data sets. All other settings remained in the default and all parameters were unlinked to allow each partition to vary independently (Mr. Bayes 3.12 manual, http://mrbayes.csit.fsu.edu/manual.php) All trees saved before convergence of the runs and stationarity of likelihood values were discarded. We determined convergence and stationarity by plotting tree number against likelihood scores for each run to find the point where the likelihood plot leveled off and began to fluctuate around a stable value using Tracer 1.4  (plots not shown). In all cases, the conservative estimate of a burnin of 25% of sampled trees was well into this plateau. The remaining 75000 trees (37500 from each run) were used to construct majority rule consensus trees and calculate posterior probabilities. All clades and support values are shown in the resulting phylogenies. All posterior probabilities are mapped on all trees. As suggested by Hulsenbeck , clades with posterior probabilities of 0.95–1.00 will be considered to be very well-supported. Clades with support values of 0.85–0.94 will be considered supported. All posterior probabilities lower than 0.85 are considered poorly supported and should be viewed with caution, but all posterior probabilities are mapped on all trees. Although taxa may appear as sister species, we can only know true sister species relationships if we have complete taxon sampling for the family.
Nomenclature and Classification
We know from the discovery of polyphyletic and paraphyletic generic groupings in Chromodoris, Glossodoris, Hypsleodoris, Mexichromis and Noumea , , that the current classification of the Chromodorididae does not reflect the evolutionary history of the group. We cannot continue to use this current classification. We will use the resulting phylogenies to propose a new classification of the chromodorid nudibranchs. The proposed new classification is based on several fundamental tnets of phylogenetic classification. Only clades are named, with two exceptions described below. Each clade contains the type species of the name-bearing clade. Exisiting, available names are utilized wherever possible to minimize the disruption to nomenclature, while simultaneously reflecting relationship. We will identify clades that include the type species of each chromodorid genus and delineate genera to minimize conflict with current classification and support recognition of interesting morphology.The translation of phylogenetic hypotheses into classifications is the best way to communicate results to a larger community, but even as the number of molecular phylogenies increases, the number of new classifications is decreasing –. The growing phylogeny/classification gap is troubling. Phylogenies are hypotheses of relationship and communicating these new hypotheses is one of the main contributions systematics can make to the scientific community.
Traditional taxonomy has obscured the patterns of diversification in the chromodorids. A new classification that properly reflects evolutionary history is required. In the new classification, we only keep existing names, for species not supported in clades if it is not disruptive to the new classification. We also hypothesize the predicted phylogenetic position of taxa that have not yet been included in the phylogenetic analysis. We used the nomenclatural standards set by the International Code of Zoological Nomenclature . Names and dates for genera, families and subfamilies we taken from Bouchet et al's review of gastropod nomenclature . Every name used is resurrected from synonymy and proposed because the type species of the genus is found in the clade. If more than one generic type species is found in the same clade, the older name has priority. In this way, the history of naming in the chromodorids will be maintained. If the gender of a species' new genus changes in the new classification, the gender of the specific epithet will be changed. Additionally, if the incorrect specific name gender has been used, the proper gender will be used in the new classification. The proposed phylogenetic naming code, the Phylocode, recommends naming clades when type species are part of the clade to be named, as we have done here, but does not use or recognize ranks as we have by using generic, subfamily and family names for clades. In many cases there is no conflict between the phylocode and traditional nomenlclature . The phylocode has not been formally adopted, so there is no official system for naming in accordance with that code. In order to maintain stability and to avoid creating names that may change with the addition of new information, we used a method advocated by Dayrat & Gosliner . This method advocates using the most inclusive known clade name as the first part of a species binomial for species that cannot be named without creating a new name. We will use the family name Chromodorididae as the name for species that would create instability if the bionomials were unchanged or if new names were given. In our proposed classification we will also include incerte sedis species in the Chromodorididae. In clades that are poorly supported (posterior probabilities below 0.85), we have used a generic name for members of those groups with the generic name placed in quotation marks. We prefer this method as an interim solution as it does not leave these taxa in taxonomic limbo and retain the use of single names for polyphyletic groups.
The sequenced COI fragment is 658 base pairs (bp) long. The edited 16s sequences are 531 bp long. The combined data sets with gaps introduced for alignment are 1189 base pairs long. All sequences are available from GenBank COI (JQ727822–JQ727914), 16s (JQ727689–JQ727821) and aligned data matrices are available upon request from the corresponding author. Excluded variable 16s regions are identified as character sets in all nexus files. Saturation was not found in the 16s fragment or the first or second positions of the COI fragment. There is slight saturation in the third position transitions in the COI data set (not shown). The third positions were included in the Bayesian analysis as the partitioning allows the parameters of this position to be estimated separately and the inclusion of the third positions did not change the resulting trees. The recommended model of evolution (AIC form Mr.Model Test) was used to set parameters in Mr.Bayes for each partition. The resulting best-fit model of evolution for each partition using the AIC selection from Mr. ModelTest ver.2  were COI 1st: GTR+G, COI 2nd: TrN+I+G, COI 3rd: GTR+I+G and 16s: GTR+I+G. These models correspond to the following settings in Mr. Bayes; all partitions set to nst = 6 and rates = invgamma except for the COI second codon position partition which was set rates = gamma.
The figured trees are the resulting consensus phylograms from the Bayesian analyses (Figures S1, S2). All posterior probabilities are shown above the branches on the Bayesian phylograms. Tree topology was not altered with the inclusion or exclusion or the 16s fragment's variable regions (See Figure S2 for comparison of trees with and without variable regions). The resulting phylogenetic hypotheses for each dataset are summarized below. We will discuss relationships in terms of posterior probabilities.
COI and 16s Combined Analysis: Including only specimens with data for both genes (Figure S1).
This data set included 164 individual chromodorids, representing 123 species, three species of Actinocyclidae, four other dorids. The outgroup was Doris kerguelensis. The data set included was 1189 bases long included gaps introduced to aid in alignment of variable regions. All bases are included. In the majority rule consensus phylogram resulting from the Bayesian analysis, the chromodorids are monophyletic (pp = 1.00). They are sister (pp = 0.98) to the monophyletic actinocyclids (pp = 1.00). Cadlinella ornatissima is sister to the rest of the chromodorids (pp = 1.00). The monophyletic Tyrinna (pp = 1.00) is poorly supported as sister to the main clade of chromodorids (pp = 0.82). The main clade of all chromodorids, except Cadlinella and Tyrinna is supported (pp = 0.85). Two clades of Noumea (both pp = 1.00) are part of a basal polytomy with the clade including the remaining chromodorid species (pp = 0.89). A well-supported clade (pp = 1.00) containing some species of Glossodoris is poorly supported at the base of the chromodorid grade. Within this main clade of chromodorids, there is one very well supported clade, which includes all species of Ceratosoma, Durvilledoris, Hypselodoris, Mexichromis, Pectenodoris, Risbecia, Thorunna and some Digidentis (pp = 1.00) Diversidoris aurantionodulosa and Noumea crocea are sister species (pp = 1.00) and poorly supported (pp = 0.78) as sister to a poorly supported clade (pp = 0.66) that includes this well-supported clade and Chromodoris alternata and Chromodoris ambiguus (pp = 1.00). There is also a poorly supported clade (or grade if the clade is collapsed) of smaller clades made up of species of Ardeadoris, Chromodoris, Diversidoris, Glossodoris, Noumea, Verconia and one species of Digidentis (pp = 0.67). Of the other 12 non-monotypic traditional genera, seven (Ceratosoma, Chromodoris, Digidentis, Glossodoris, Hypselodoris, Mexichromis, Noumea) are non-monophyletic and three (Durvilledoris, Pectenodoris, Risbecia) are monophyletic but render another genus paraphyletic. Both Ardeadoris and Thorunna are made paraphyletic by nested members of other genera (Noumea, Glossodoris-within Ardeadoris and Digidentis-within Thorunna). There are three species and five clades of eastern Pacific and/or Atlantic species.
COI and 16s Combined Analysis: Including all specimens (Figure S2).
This data set included 244 individual chromodorids, representing 157 species, four species of Actinocyclidae, four other dorids. The outgroup was Doris kerguelensis.
The complete data set included 1189 bases. The chromodorids are monophyletic (pp = 0.94). They are sister to the monophyletic actinocyclids (pp = 0.98). A clade including both species of Cadlinella is sister to the rest of the chromodorids (pp = 1.00). The monophyletic Tyrinna (pp = 1.00) is poorly supported as sister to the main clade (pp = 0.83). There are two clades containing species of Noumea and the one species of Verconia (pp = 1.00 and pp = 1.00) that form a polytomy with the clade of all of the remaining chromodorids (pp = 0.85). Within the main clade of chromodorids, there is one very well supported clade, which includes all species of Ceratosoma, Durvilledoris, Hypselodoris, Mexichromis, Pectenodoris, Risbecia, Thorunna and some Digidentis (pp = 1.00) and a grade of clades of species Ardeadoris, Chromodoris, Diversidoris, Glossodoris, Noumea, Verconia and one species of Digidentis. Of the other 12 non-monotypic traditional genera, seven (Ceratosoma, Chromodoris, Digidentis, Glossodoris, Hypselodoris, Mexichromis, Noumea) are non-monophyletic and three (Durvilledoris, Pectenodoris, Risbecia) are monophyletic but render another genus paraphyletic. Both Ardeadoris and Thorunna are made paraphyletic by nested members of other genera (Noumea, Glossodoris and Digidentis) (Figure 3). More detailed results found within each clade will be discussed below. There are three individual species and five clades of eastern Pacific or Atlantic species (Figure 4). The data set without variable regions included 1108 bases. There are only slight changes to the tree topology, including slight losses of support and changes to branching pattern in the species relationships within four clades containing species of Noumea, Glossodoris, Chromodoris and Thorunna. Additionally, some clades are more well-supported in the phylogram without variable regions, most notably there is some support for the two Noumea clades as sisters. All of these differences are mapped in mirrored trees (Figure S2).
Tree is the same Bayesian phylogram as figured in S3A. All specimens, both genes and all characters included.
Tree is the same Bayesian phylogram as figured in S3A. All specimens, both genes and all characters included. Blue = Indo-Pacific, Red = Atlantic and Caribbean, Gold = Eastern Pacific, Green = Sister group, Black = outgroups. Dark grey = Solely eastern Pacific and Atlantic clades. Light grey = Primarily Indo-Pacific clades with eastern Pacific members.
New Classification of the Chromodorididae (Table S2)
Our classification is based on our COI and 16s combined phylogeny with all specimens included (Figure S2A). The older name that describes this clade, Ceratosomatidae Gray 1857  was declared nomen oblitum under Art. 23.9 of the International Code of Zoological Nomenclature . Even though it is older than the name in current usage, Chromodorididae, it had not been used in over fifty years. In the phylogeny of the chromodorid nudibranchs, there are five basal clades: Cadlinella, Tyrinna, two clades made up of some species of Noumea and Verconia and one clade made up of some species of Glossodoris. There is one, main, well-supported clade including species of Ceratosoma, Hypselodoris, and grade of clades. We will briefly introduce each clade and its member species in the context of a new classification for chromodorid nudibranchs (Figure 5, S2A, S3).
Tree is the same Bayesian phylogram as figured in S3A. All specimens, both genes and all characters included.
Of the 16 genera in the current chromodorid classficiation, only two, Cadlinella and Tyrinna, retain the same membership in our new classification. Nine more generic names are retained; Ardeadoris, Ceratosoma, Chromodoris, Diversidoris, Glossodoris, Hypselodoris, Mexichromis, Noumea and Thorunna but their membership has changed. Five names are synonymized with the names listed above: Digidentis, Durvilledoris, Pectenodoris, Risbecia and Verconia. Five older names: Doriprismatica, Felimare, Felimida, Goniobranchus and Miamira, have been rescued from synonymy and are used to describe clades of species previously included in different genera. There are 17 generic names used in our new classification, each of which will be detailed below and in Figure S3. Of these, 13 are very well supported with posterior probabilities ≥0.95. Mexichromis is supported with a posterior probability of 0.87 when variable positions are included and 0.95 when they are excluded. Noumea consists of two separate clades (both pp = 1.00) that are poorly supported as a combined clade in the analysis when variable positions are included (pp = 0.61). Although, this support is not sufficient, all of the species in both of these clades are currently named Noumea and will retain this name in order to maintain stability. Doriprismatica is extremely poorly-supported pp = 0.64 with variable positions included and well-supported (pp = 0.92) in the analysis with variable positions excluded (Figures S2, S3). We do not consider this level of support sufficient to definitively name this clade, but because continuing to use the current name, Chromodoris, would add greater confusion (as that name represents a different, well-supported monophyletic group) we will premilinarily name these species ‘Doriprismatica’. Similarly, a group of species some of which are currently classified as Chromodoris and some as Glossodoris form a polytomy together with other well supported clades. As these species need a name, but lack appropriate support, they cannot be named definitively. We will preliminaryily name these species ‘Felimida’ . Naming these clades is much more stable than using names that now represent other well-defined and well supported clades. These names are hypotheses; with more data the relationships of members of these clades will likely become better resolved. The completed classisifiction is listed in Table S2.
Chromodorididae Bergh 1891 
Cadlinella Thiele, 1931 .
Type species: Cadlina ornatissima Risbec, 1928  (by monotypy)
The two species of Cadlinella included here, Cadlinella ornatissima and Cadlinellla subornatissima form a clade and are sister to the rest of the chromodorid species (pp = 1.00). These findings support previous results ,  and Rudman's evolutionary scenario , . The widespread Indo-Pacific genus, Cadlinella is an enigmatic taxon. It has at different times been considered it own separate family , a part of the Cadlininae  and a member of the Chromodoridinae/Chromodorididae , .
Tyrinna Bergh, 1898 .
Type species: Tyrinna nobilis Bergh, 1898  (by monotypy)
The only two species of Tyrinna: T. evelinae and T. nobilis are included here. Tyrinna is always monophyletic (pp = 1.00). After the split from Cadlinella, this clade is poorly-supported as the sister group to the main group of chromodorids (pp = 0.83). Rudman  suggested that Tyrinna, Cadlinella and Cadlina form a basal grade of primitive chromodorids. Cadlina had been shown not to be a chromodorid , but our results support Rudman's suggestion that Tyrinna and Cadlinella are basal to the rest of the chromodorids. Muniain et al  and Schrödl and Millen  extensively reviewed the morphology of the two species in this clade.
Noumea Risbec, 1928 .
Type species: Noumea romeri (by subsequent designation Baba, 1937 )
Verconia Pruvot-Fol, 1931 
Type species: Albania? verconis Basedow and Hedley, 1905  (by monotypy)
Verconia verconis is well supported as part of a clade that includes N. haliclona, N. laboutei, N.romeri and N. simplex (pp = 1.00). Noumea varians, N. purpurea and N. norba form a well-supported clade (pp = 1.00) that is not part of a name bearing clade, but is one branch of the polytomy that includes the ‘Noumea sensu stricto’ and the branch leading to the rest of the family (pp = 0.88). The monotypic genus Verconia is nested within the Noumea clade as suggested by Rudman  and weakly supported as the sister species to another South Australian species, N. haliclona, as found in the preliminary results shown by Turner & Wilson .
Glossodoris Ehrenberg, 1831 .
Type species: Doris xantholeuca Ehrenberg, 1831  = G. pallida (by subsequent designation)
The Glossodoris clade (pp = 1.00) includes species G. pallida and G. rufomargninata. In an important, but often overlooked detailed examination of the relationships of the species classified in the genus Glossodoris, Rudman identified five subgroups of this genus based on morphology . The species in this Glossodoris clade were considered by Rudman  to be members of the ‘Glossodoris pallida subgroup’. This clade also includes two species he did not include in any subgroup, G. cincta and G. hikuerenesis.
Goniobranchus Pease, 1866 .
Type species: Goniobranchus vibrata Pease, 1866  (by subsequent designation)
This clade includes all of the Indo-Pacific species of Chromodoris that are not part of the black-lined, planar egg mass clade (pp = 1.00), except Chromodoris alternata and Chromodoris ambiguus. This phylogeny is the first to find definitive support for a clade of chromodorids, first suggested by Wilson  and Turner and Wilson  known to lay egg masses with extra-capsular yolk. When Pease designated Doris vibrata as the type species for the new genus Goniobranchus, he should have changed the ending of vibrata to vibratus to reflect the masculine gender of the –us ending. We have made that correction here and changed the gender of all of the species names that require changing (names derived from adjectives) in Goniobranchus.
‘Doriprismatica’ d'Orbigny, 1839 .
Species included in the Glossodoris atromarginata subgroup  are recovered in this clade, with the addition of G. sedna and Digidentis kulonba (pp = 0.95).
‘Felimida’ Marcus, 1971 .
Type species: Felimida sphoni Marcus, 1971  (by monotypy)
This name will be used for all eastern Pacific and Atlantic species of Chromodoris and Glossodoris (except Glossodoris sedna). These species form a polytomy including Glossodoris baumanni and three clades of Atlantic and Eastern Pacific chromodorids.
Chromodoris clenchi, C. norrisi and C. sphoni (pp = 1.00)
Glossodoris dalli and G. edmundsi (pp = 1.00)
Chromodoris krohni, C. luteorosea and C. purpurea (pp = 0.78)
These exclusively eastern Pacific and Atlantic clades do not form a monophyletic group, but we will provisionally name all of these species ‘Felimida’. This is the most conservative choice, the choice that requires the fewest name changes and is the least disruptive pending further information and broader taxon sampling.
Ardeadoris Rudman, 1984 .
Type species: Ardeadoris egretta Rudman, 1984  (by original designation)
The Ardeadoris clade includes both species of Ardeadoris: A. egretta and A. scottjohnsoni, five species of Glossodoris (G. averni, G. pullata, G. rubroannulata, G. tomsmithi and Glossodoris undaurum) and Noumea angustolutea (pp = 1.00). . Based on their analysis, Turner and Wilson  suggested that with more sampling it would be come clear if Ardeadoris should be synonmized with Glossodoris. By sampling more broadly within the family, we found the converse. Four species of Glossodoris and Noumea angustolutea need to be included within Ardeardoris because they are strongly supported as part of the clade including Ardeadoris egretta and not the type species of Glossodoris. Three of the species, G. averni, G. undaurum and G. rubroannulata, found in this clade were part of Rudman's Glossodoris sedna subgroup 
Chromodoris Alder & Hanncock, 1855 .
Type species: Doris magnifica Quoy and Gaimard, 1832  (by original designation)
This clade includes all of the black-lined species of Chromodoris and Chromodoris aspersa (pp = 1.00). This clade was identified by, both Wilson & Lee  and Turner & Wilson , as the planar spawning or black-lined Chromodoris clade. All of the members of this clade lay flat egg masses.
Diversidoris Rudman, 1987 .
Type species: Diversidoris aurantionodulosa Rudman, 1987  (by original designation).
The Diversidoris includes, Diversidoris aurantionodulosa, two yellow species of Noumea, N. crocea and N. flava, and a new species from Moorea, French Polynesia-Chromodoridae BioCode 2937 (pp = 0.95).
Miamirinae Bergh 1891 
The Miamirinae clade includes all of the species currently classified as Ceratosoma, Durvilledoris, Hypselodoris, Mexichromis, Pectenodoris, Risbecia, Thorunna and two species of Digidentis (pp = 1.00)
This clade was first predicted by Rudman  based on morphological similarities and then confirmed by Rudman & Berquist's  finding that all of the species in this clade feed exclusively on sponges of the family Dysideaidae, although they assumed all of the genera to be monophyletic. Miamirinae Bergh 1891 is the oldest appropriate and available subfamily or family name for this clade. The remaining six genera; Miamira, Ceratosoma, Felimare, Mexichromis, Thorunna and Hypselodoris make up the Miamirinae.
Miamira Bergh, 1874 .
Type species: Miamira nobilis Bergh, 1874  (by monotypy)
The Miamira clade includes the following species (as currently classified) Ceratosoma alleni, Ceratosoma magnificum, Ceratosoma miamiranum, Ceratosoma sinuatum. Miamira is part of a grade with Ceratosoma. The morphological phylogeny of species of Ceratosoma and classified as Miamira and Orodoris, that was used as justification for their synonomy, predicted a sister group relationship between species of Miamira and Ceratosoma alleni . Our results confirm that C. alleni is more closely related to species of Miamira, but do not find support for synonymy of Miamira and Ceratosoma. Although, it is possible this relationship will be recovered with further sampling and by including molecular markers that will help resolve basal branches on the phylogeny.
Ceratosoma Adams and Reeve, 1850 .
Type species: Ceratosoma cornigerum Adam and Reeve, 1850  (by monotypy)
The Ceratosoma clade includes C. amoenum, C. gracillimum, C. ingozi, C. tenue, C. trilobatum and a new species. (pp = 1.00)
Felimare Marcus and Marcus, 1967 .
Type species: Felimare bayeri Marcus and Marcus, 1967  (by monotypy)
The Felimare clade includes all eastern Pacific, Atlantic and Mediterranean species of Hypselodoris and two species of Mexichromis, M. porterae and M. kempfi from the eastern Pacific and Caribbean respectively (pp = 1.00). Both Gosliner and Johnson  and Alejandrino and Valdés  hypothesized a sister group relationship between the Indo-Pacific and eastern Pacific/Atlantic species of Hypselodoris. Turner and Wilson  did not recover that relationship, but instead found the same relationships shown here.
Mexichromis Bertsch, 1977 .
Type species: Chromodoris antonii Bertsch, 1976  (by original designation)
This clade includes the type species of Mexichromis, M. antonii, known only from the eastern Pacific and the three included species of Durvilledoris, D. lemniscata, D. pusilla and D. similaris, the two described species of Pectenodoris, P. aurora and P. trilineata and all of the Indo-Pacific species currently considered Mexichromis, M. festiva, M. macropus, M. mariei and M. mutituberculata (pp = 1.00). There are two well-supported clades within the Mexichromis clade. The clade including Mexichromis antonii and the species of Durvilledoris is sister to the clade including Pectenodoris and Indo- Pacific Mexichromis. These clades could be given two names, but it is much less disruptive and confusing to maintain the name Mexichromis for all clade members. The clade including P. aurora and P. trilineata can be called the ‘Pectenodoris’ clade of Mexichromis.
Thorunna Bergh, 1878 .
Type species: Thorunna furtiva Bergh, 1878  (by monotypy)
The Thorunna clade includes all species of Thorunna and two species of Digidentis, D. arbutus and D. perplexa. All of species currently classified as Thorunna are found in the Indo-Pacific and the species of Digidentis are limited to southern Australia. As suggested by Rudman , the only species within Thorunna with mantle glands, T. australis and the species of Digidentis (all of which have mantle glands) form a clade.
Hypselodoris Stimpson, 1855 .
Type species: Goniodoris obscura Stimpson, 1855  (by monotypy)
This clade includes all of the Indo-Pacific species of Hypeslodoris and Risbecia (pp = 1.00).
Species of Risbecia s.s forms a well-supported clade nested within Hypselodoris and can be referred to as the Risbecia clade of Hypselodoris. Risbecia aplogema is not part of this Risbecia clade and was previously considered a species of Hypselodoris. Including all of the members of the Risbecia and Hypselodoris bullocki clade in Risbecia is not an option because this would render Hypsleodoris paraphyletic. The second clade includes, H. bennetti, H, maritima, H. bertschi, H. paulinae, H. kaname, H. bollandi, H. obscura, H. infucata, H. zephrya and one or two new species. The third clade includes H. reidi, H. krakatoa, H. jacksoni and one new species. This clade was also recovered in Gosliner & Johnson's  morphological phylogeny of Hypselodoris.
Chromodoris alternata and Chromodoris ambiguus.
The enigmatic south Australian species, Chromodoris alternata and C. ambiguus are very different than other chromodorids. They are two of the five chromodorid species with a plesiomorphic serial reproductive system (C. loringi, C.thompsoni, C. woddwardae) , , . All five of these species are found only in southeastern Australia. These species were found to be more closely related to Cadlina than Chromodoris by Wilson & Lee , but as part of the chromodorid grade in Turner & Wilson . Clearly further work on this group and its relationship to all cryptobranchs is needed. The addition of specimens of C. loringi, C. thompsoni and C. woodwardae , , , the only other chromodorid species known to have a serial reproductive system may help solve this problem. These two species are always each other's closest relatives and are sister to the rest of the Miamirainae in the all analyses. As suggested by Dayrat & Gosliner  they should be considered Chromodorididae, because they are not included in a named clade. Until the ambiguity of the relationship of these taxa to other chromodorids can be resolved, they should be considered Chromodorididae alternata and Chromodorididae ambiguous.
Phylogeny of the chromodorid nudibranchs.
The primary goal of this study was to generate a phylogeny of the chromodorid nudibranchs and present a classification that accurately reflects the evolutionary history of this group. We have included mitochondrial DNA sequence data for 157 chromodorid species, more than double previous sampling and making this the largest species level phylogeny of nudibranchs ever published. We included the type species of every genus; complete sampling of every described species for five of the sixteen genera (Ardeadoris, Diversidoris, Pectenodoris, Tyrinna and Verconia) and more than half of the species in every other genus. With this sampling, we were able to test the monophyly of all the chromodorid genera. Both Rudman's ‘Chromodoris subgroup’ including species of Ardeadoris, Chromodoris, Glossodoris, Noumea (only some species, other species currently considered Noumea are also found in the Ardeadoris and Diversidoris clades) and his ‘Hypselodoris subgroup’ with species of Ceratosoma, Hypselodoris, Mexichromis and Thorunna were recovered. Although these subgroups can be observed, none of the chromodorid genera is monophyletic, except Cadlinella and Tyrinna (see above). In every case, a genus either is polyphyletic or it is nested within another genus; and therefore it makes another genus paraphyletic. This result at once illuminates both the difficulty of delineating natural groups in very diverse and homoplastic clades and the insight that can be gained from systematic reviews, like Rudman's  review of the family. Rudman  was able to discern the two main groups of chromodorids without a phylogenetic analysis or molecular data. At a large enough scale, it is possible to sort out synapomorphies, but on a smaller scale, homoplasy muddies the waters. The problem of homoplasy confusing taxonomy and systematics in nudibranchs has been explored , but is not generally mentioned in the description of new genera. This is largely a result of the fact that the majority of new taxa have been described without a phylogenetic hypotheses and similar morphology is rarely discussed outside the possibility of close relationship rather than being a result of homoplasy. Traditionally, new genera were often erected on the basis of a single evolutionary novelty. Many of thes attributes prove to be autapomorphies in these taxa and do not consitiute a basis for establishing these genera as clades when subjected to phylogenetic analysis.
In future contribtuions, we will work out synapomorphies for the clades identified here, but because of the amount of homoplasy and number of incomplete descriptions, this is a huge undertaking and not appropriate here.
Monophyly of the Chromodorididae.
Bergh  was the first to suggest a separate taxonomic rank for the chromodorid nudibranchs. Johnson  showed that the Chromodorididae are only monophyletic if Cadlina is removed from the family, as Cadlina is more closely related to Aldisa and other dorid nudibranchs. We expanded on these preliminary results and confirmed the monophyly of the chromodorids in analyses without Cadlina and without including as many dorid species. Gosliner and Johnson  reviewed the genus Hallaxa and presented a morphological phylogeny of Hallaxa and Actinocylcus. They hypothesized that the semi-serial reproductive system found in species of Hallaxa and Actinocyclus and all chromodorids is a synapomorphy that unites the two groups. The chromodorid nudibranchs are monophyletic in every analysis, as is their sister group relationship with Actinocyclidae. These analyses confirm the hypothesis of Gosliner & Johnson  and the preliminary findings of Johnson  and confirm the utility of this morphological synapomorphy.
Most previous work has assumed monophyly of chromodorid genera; subsequently work on the natural history of chromodorids has used genera as de facto hypotheses of relationship. Genera should only be used in this way if they are known to be monophyletic through phylogenetic analysis. There have been two classes of ‘naming problems’ in the nomenclatural history of the chromodorid nudibranchs. The first can be described as the novelty problem (as described above), when unique or ‘unclassifiable’ species were discovered new genera we created to contain them , . And the second, the ‘catch-all’ problem, new species were assigned to large genera with the widest definitions . Considering these genera as evolutionary units at the broad scale may not lead to mistakes, but at a finer scale, we may be missing the true origins of novelty or by grouping things that are superficially similar together we may miss repeated origins of diversity (convergence). Turner & Wilson  were the first to truly test generic monophyly in more than one chromodorid genus (See Figure 1 for previous phylogenetic hypotheses). They found evidence for the non-monophyly of most chromodorid genera. The only genera they found to be monophyletic were Digidentis (pp = 1.00), and Durvilledoris (pp = 1.00). Of the genera they could test, they found Chromodoris, Glossodoris, Hypselodoris and Mexichromis to be paraphyletic or polyphyletic. They also found Risbecia to be monophyletic, but nested within Hypselodoris. They used their findings as evidence for the ‘polyphyly of widespread genera’ or species currently classified in different genera found in one ocean basin more closely related to each other than to their congeners found in other oceans. For example, they found Mexichromis porterae, a species known only from the eastern Pacific, to be more closely related to species of Hypselodoris from the eastern Pacific than either were to Mexichromis or Hypsleodoris species found in the Indo-Pacific. This finding tells us something new, but it actually does not tell us much about biogeography, because the genera they discussed were not monophyltic entities. They uncovered a taxonomic problem, not a biogeographic or biological one. Their results actually confirm the taxonomic confusion the authors of most of the species they sampled expressed when faced with choosing a generic placement for new taxa. In fact, all of these authors established new genera to account for differences they found, Mexichromis, Felimare, Felimida, Digidentis, Ardeadoris, Durvilledoris etc , , , . It was primarily by changes to the generic placement by subsequent authors, synonymy of, and the addition of taxa to, newly created genera that lead to the ‘polyphyly across the oceans’ , . The phylogeny presented here allows informed exploration of the taxonomic and nomenclatural history of the chromodorid nudibranchs. The clades we recovered in this molecular phylogeny are even more interesting when we use the taxonomic history, in the form of older generic names attached to type species, as a map of the discovery of the great diversity of this group. Our studies indicate that when monophyletic units are recognized, there is strong biogeographical signal rather than “polyphyly across oceans”.
In summary, with the most comprehensive sampling of chromodorid species to date, we confirmed that the chromodorids are monophyletic and are sister to the monophyletic actinocyclids. We also found that the majority, 12/14 non-monotypic traditional genera, were not monophyletic or make another clade paraphyletc. Seven traditional genera, Ceratosoma, Chromodoris, Digidentis, Glossodoris, Hypselodoris, Mexichromis, Noumea are non-monophyletic and three (Durvilledoris, Pectenodoris, Risbecia) are monophyletic but render another genus paraphyletic. Both Ardeadoris and Thorunna are made paraphyletic by nested members of other genera (Noumea, Glossodoris and Digidentis). The two monotypic genera, Diversidoris and Verconia are nested within clades. Only Tyrinna and Cadlinella are monophyletic and without disruption to any other clades (Figure 3, S1, S2). The classification proposed here and discussed at length above renames clades and is more consisitent with evolutionary history (Figure S3).
Biogeography in light of new classification.
The most speciose chromodorid genera: Chromodoris, Glossodoris and Hypselodoris were originally created to describe Indo-Pacific species. It wasn't until some time after these names were created that previously described, similar, brightly colored cryptobranch dorid species found in the eastern Pacific, western Atlantic and Mediterranean were added to these genera , , , , , . In Mexichromis the opposite is true. The type species, Mexichromis antonii, was described from the eastern Pacific and Indo-Pacific species were included later included in this genus , . Other eastern Pacific “Mexichromis” are shown here to belong to Felimare.
This new classification clarifies our view of biogeographic patterns in the chromodorid nudibranchs. Instead of taxonomy obscuring patterns of diversification in this group, this taxonomy reflects and reinforces evolutionary history. It gives us a much better framework for exploring evolutionary questions.
The majority of chromodorid nudibranchs are found in the Indo-Pacific, but there are three individual species and five clades of solely Atlantic and/or eastern Pacific species (Figure 4). The sister group to the rest of the chromodorids, Cadlinella is found only in the Indo-Pacific, while the sister to the Chromodorididae, the Actinocyclidae is found in most temperate and tropical waters. Although there are other possibly scenarios, such as trans-Pacific dispersal and migration around Africa, the pattern uncovered here, strongly supports the simplest hypothesis that the chromodorids diversified rapidly from the tropical Tethyan Realm. This pattern has been found in other gastropod groups – (Figure 4). The chromodorids were likely widely distributed and different lineages diversified in isolation following vicariant events. This scenario is further supported by the fact that Goniobranchis is sister to ‘Doriprismatica’ and its closest realitves and that all the memebers of Goniobranchus are Indo-Pacific. Also in this scenario, the specimens identified as D. sedna from the Atlantic and eastern Pacific appear to be distinct species as indicated by COI pairwise distances of 11.7–11.0% between eastern Pacific and Altantic specimens while the three eastern Pacific specimens are 0–0.7% different from each other. This scenario clearly supports vicariance between the Indo-Pacific and eastern Pacific and Atlantic preceding the vicariance between the eastern Pacific and Atlantic. In the main chromodorid grade of clades there are two individual species and three clades that are exclusively Atlantic and/or eastern Pacific. Specimens identified as ‘Doriprismatica’ sedna found both in the eastern Pacific and the western Atlantic, are always sisters and are nested within a clade of exclusively Indo-Pacific species. This is most likely a radiation into the eastern Pacific and Atlantic from the Indo-Pacific. The remainder of the Atlantic and eastern Pacific species, not included in the Miamirinae, are part of a polytomy including five clades, three containing only eastern Pacific and Atlantic species of ‘Felimida’ and the Indo- Pacific Ardeadoris and Chromodoris clades. ‘Felimida’ baumanni, found in the eastern Pacific, is also part of this polytomy. The relationships in this grade need to be examined more closely with the addition of more specimens and more genes.
Relationships within the Miramirinae clade are more resolved. There are two clades that include eastern Pacific and Atlantic species. The Felimare clade is exclusively eastern Pacific and Atlantic. There are two eastern Pacific and Atlantic splits in this clade, the eastern Pacific F. porterae and Caribbean F. kempfi are potentially geminate species and are sister to a larger clade of eastern Pacific, Caribbean and eastern Atlantic Felimare species. More sampling is needed in this clade to further untangle the emergent biogeographic patterns within Felimare. Additionally, within the Miamirinae, the eastern Pacific species, Mexichromis antonii, is sister to the exclusively Indo-Pacific M. lemniscata, M. pusilla and M. similaris and this clade is sister to the rest of the Indo- Pacific Mexichromis. Within the Miamirinae, it appears that there had been more than one dispersal event from the Indo-Pacific, to eastern Pacific and Atlantic. Ceratosoma and Miamira species are only found in the Indo-Pacific and adjacent temperate regions. The sister taxon, the clade including: Felimare, Mexichromis, Thorunna and Hypselodoris, has a wider distribution. Within this sister taxon, Felilmare is exclusively eastern Pacific and Atlantic while its sister species in its sister taxon Mexichromis, Thorunna and Hypselodoris are almost exclusively Indo-Pacific and adjacent temperate regions. Mexichromis antonii, which is found in the eastern Pacific, is the only speices in this clade not found in the Indo-Pacific or adjacent regions. Thus, Felimare and Mexichromis antonii represent two distrinct invasions of the eastern Pacific from the Indo-Pacific.
We hope the work presented here will serve as a starting point for further research into the evolutionary history of the chromodorid nudibranchs. This phylogeny is based only on mitochondrial genes, one of our first next steps will be to include sequences from nuclear genes for all of the species included here. The addition of more slowly evolving unlinked markers should help resolve some poorly supported node at the base of the phylogenies presented here and will add a separate line of evidence to this hypothesis. Addtionally, morphological synapomorphies need to be found for the clades recoverd in this phyogeny. There is still much work to do in order to untangle the evolutionary history of this group. This phylogeny and classification is a start that will allow us to use names that represent monophyetic groups as the starting point for future discovery.
The resulting classification can be used to address questions of interest to a much broader community. A robust phylogeny and corresponding revised classification are necessary to conduct comparative studies in the Chromodorididae. Evolutionary studies of trophic specialization, color patterns and secondary metabolites, for example, will be much more robust by comparing monophyletic units that are clearly named within a new classification. We have pointed out and highlighted many areas of future research. As we detail above, there are many taxonomic, nomenclatural and species delineation problems that still require refinement within the chromodorid nudibranchs. These questions can best be answered with a detailed examination of morphology together with molecular data. The molecular data needs to be rooted in sound identifications and definitions that are always based on morphology. This phylogeny does not answer all of these issues but it will serve as a framework to more effectively tackle these questions. Our classification will serve as a more refined basis for other evolutionary biologists, ecologists and natural products chemists. Their results will be more informative in light of a classification based on evolutionary history rather than one based on untested hypotheses. Phylogenetic systematics provides a rigorous and repeatable methodology that permits iterative approximations of relationship and our understanding of this diverse and biologically intriguing group of organisms is enhanced by these studies.
The electronic version of this document does not represent a published work according to the International Code of Zoological Nomenclature (ICZN), and hence the nomenclatural acts contained in the electronic version are not available under that Code from the electronic edition. Therefore, a separate edition of this document was produced by a method that assures numerous identical and durable copies, and those copies were simultaneously obtainable (from the publication date noted on the first page of this article) for the purpose of providing a public and permanent scientific record, in accordance with Article 8.1 of the Code. The separate print-only edition is available on request from PLoS by sending a request to PLoS ONE, Public Library of Science, 1160 Battery Street, Suite 100, San Francisco, CA 94111, USA along with a check for $10 (to cover printing and postage) payable to “Public Library of Science”.
Table of Specimens Used in this Study. Specimens used in this study listed by Family. The names in this table reflect current classification not proposed classification (new names are listed in the text). Abbreviations are as follows: CASIZ = California Academy of Sciences, SAM = South Australian Museum, WAM = Western Australian Museum, AM = Australian Museum, ZSM = Zoologische Staatssammlung München, SIO-BIC = Scripps Institute of Oceanography, BioCode = Moorea BioCode Project.
New Classification of the Chromodorididae with synonyms. Generic names and type species in bold and the most recent genus membership follows. Listing order follows phylogeny.
Bayesian consensus phylogram including all specimens with data for both genes. Posterior probabilities are listed above branches. Doris kerguelensis is the outgroup. This phylogram is the consensus of 50,000,000 generations with trees sampled every 1000 generations with a burnin of 25%. Data was partitioned by gene and by codon position.
Bayesian consensus phylograms including all specimens. A. Phylogram resulting from the inclusion of all characters B. Phylogram resulting from excluding hard to align characters. Doris kerguelensis is the outgroup. These phylograms are the consensuses of 50,000,000 generations with trees sampled every 1000 generations with a burnin of 25%. Data was partitioned by gene and by codon position. Dotted lined indicate areas of disagreement.
Cladogram drawn from Bayesian tree with all specimens, both genes and all characters included. (S3A). Posterior probabilities are listed above branches. New generic names are used in the tree. Types are in bold. Photos of selected specimens are for reference and to show the range of diversity in each genus and the family. From top to bottom: Cadlinella Thiele, 1931 Cadlinella ornatissima, CASIZ 159381 Mooloolaba, Australia, Robert Mann. Tyrinna Bergh, 1898 Tyrinna evelinae, Costa Rica, TMG. Noumea Risbec, 1928: Left Noumea romeri, CASIZ 159896, Mooloolaba, Australia Robert Mann. Right Noumea norba CASIZ 156661, Philippines, TMG. Glossodoris Ehrenberg, 1831 Upper Glossodoris cincta, CASIZ 158809, Philippines, Ángel Valdés. Lower Glossodoris pallida, CASIZ 157871, Philippines, TMG. Goniobranchus Pease, 1866 Upper Goniobranchus vibratus CASIZ 175564, Hawaii, USA, Cory Pittman. Middle Goniobranchus fidelis, CASIZ 175426, Philippines, Ángel Valdés. Lower Goniobranchus reticualtus CASIZ 156921 TMG. ‘Doriprismatica’ d'Orbigny, 1839 Upper ‘Doriprismatica’ atromarginata, CASIZ 177237, Philippines, TMG. Lower ‘Doriprismatica’ stellata, Papua New Guinea, TMG. ‘Felimida’ Marcus, 1971. Left Felimida sphoni CASIZ 175431, Costa Rica, TMG. Felimida norrisi, Baja California, TMG. Ardeadoris Rudman, 1984. Upper Ardeadoris egretta, CASIZ 157481, TMG. Lower Ardeadoris angustolutea, CASIZ 121068, Marshall Islands,Scott Johnson. Chromodoris Alder & Hanncock 1855. Upper Chromodoris magnifica, CASIZ 157027, Philippines, TMG. Lower Chromodoris colemani CASIZ 158766, Philippines, TMG. Diversidoris Rudman, 1987. Left Diversidoris crocea, Philippines, TMG. Right Diversidoris auriantonodulosa, New South Wales, Australia, Denis Riek. Miamira Bergh, 1874. Left Miamira alleni, CASIZ 180411, Philippines, TMG. Right Miamira sinuata, CASIZ 166764, Okinawa, Robert Bolland. Ceratosoma Adams & Reeve, 1850. Left Ceratosoma cf. tenue, CASIZ 156077, Mooloolaba, Australia, Shireen Fahey. Right Ceratosoma trilobatum, CASIZ 173451, Madagascar, TMG. Felimare Marcus and Marcus, 1967. Upper Felimare bayeri CASIZ 175461, Bocas del Toro, Panama, Shireen Fahey. Lower Felimare agassizii, Baja California, Mexico, TMG. Mexichromis Bertsch, 1977. Upper Mexichromis antonii, CASIZ 175436, Costa Rica, TMG. Lower Mexichromis trilineata, Philippines, TMG. Thorunna Bergh, 1878. Upper Thorunna florens, CASIZ 177094, Vanuatu, Yolanda Camacho. Lower Thorunna furtiva, CASIZ 175729, Malaysia, TMG. Hypselodoris Stimpson, 1855. Upper Hypselodoris zephyra, Philippines, TMG. Middle Hypselodoris capensis, South Africa, TMG. Lower Hypselodoris obscura, Jervis Bay, NSW, Australia, Sue Newson.
Ángel Valdés, Don Potts, Nudibranch Central at CAS, especially Marta Pola, Yolanda Camacho, Jamie Chan and Benoit Dayrat, the Potts lab UCSC, the Center for Comparative Genomics at CAS, especially Josh Hallas, Boni Cruz, Susie Lockhart, Monica Medina, Kristy Deiner, Kristen Cella, Alessandra Lopez, Anna Sellas and Brian Simison. A work of this scope is a true collaboration and would have been impossible with the dedication and knowledge of: Cory Pittman, Pauline Fiene, Robert Bolland, Scott Johnson, INBIO and Yolanda Camacho, Lucas Cervera, Larry Harris, Alicia Hermosillo, Yvonne Valles, Benoit Dayrat, Mike Miller, Hans Bertsch, Marta Pola, Goncalo Caladao, Shireen Fahey, Patrick Krug, Manuel Malquias, Correy Whisson of WAM, Lindsay Groves of LACM, Moorea BioCode Project, Philippe Bouchet and Census of Marine Life-Pangolo and Santo expeditions.
Conceived and designed the experiments: RFJ TMG. Performed the experiments: RFJ. Analyzed the data: RFJ. Contributed reagents/materials/analysis tools: TMG. Wrote the paper: RFJ. Conceptualized and wrote the new classification: TMG RFJ.
- 1. Edmunds M (1981) Opishtobranchiate Mollusca from Ghana: Chromodorididae. Zoological Journal of the Linnean Society 71: 175–201.
- 2. García-Gómez JC, Cimino G, Medina A (1990) Studies on the defensive behavior of Hypselodoris species (Gastropoda: Nudibranchia) Ultrastructure and chemical analysis of mantle dermal formations (MDFs). Marine Biology 106: 245–250.
- 3. García-Gómez JC, Medina A, Covenas R (1991) Study of the anatomy and histology of the mantle dermal formations (MDFs) of Chromodoris and Hypselodoris (Opisthobranchia: Chromodorididae). Malacologia 32(2): 41–46.
- 4. Lüttmann K, Anthes N, D'Souza TG, Riss S, Michiels NK (2006) Population size estimate of a reef-flat aggregation of Chromodoris annulata (Opisthobranchia, Chromodorididae). Journal of Molluscan Studies 72(2): 214–216.
- 5. Rudman WB, Berquist PR (2007) A review of feeding specificity in the sponge-feeding Chromodorididae (Nudibranchia: Mollusca). Molluscan Research 27(2): 60–88.
- 6. Faulkner DJ, Ghiselin MT (1983) Chemical defense and evolutionary ecology of dorid nudibranchs and some other opisthobranch gastropods. Marine Ecology Progress Series 13: 295–301.
- 7. Avila C (1995) Natural products of opisthobranch molluscs: a biological review. Oceanography and Marine Biology Annual Review 33: 487–559.
- 8. Cimino G, Fontana A, Gavagnin M (1999) Marine opisthobranch molluscs: chemistry and ecology in sacoglossans and dorids. Current Organic Chemistry 3: 327–372.
- 9. Mollo E, Gavagnin M, Carbone M, Guo Y, Cimino G (2005) Chemical studies on Indopacific Ceratosoma nudibranchs illuminate the protective role of their dorsal horn. Chemoecology 15: 31–36.
- 10. Cimino G, Ghiselin MT (2009) Chemical Defense and the Evolution of Opistobranch Gastropods. Proceedings of the California Academy of Sciences 60(10): 175–422.
- 11. Gosliner TM, Behrens D (1990) Special resemblance, aposematic coloration and mimicry in opisthobranch gastropods. In: Wicksten M, editor. Symposium on the Adaptive Significance of Color in Invertebrates. Sponsored by the American Society of Zoologists. College Station: Texas A. & M. University Press. pp. 127–138.
- 12. Rudman WB (1991) Purpose in pattern: the evolution of colour in chromodorid nudibranchs. Journal of Molluscan Studies 57: 5–21.
- 13. Gosliner TM, Johnson RF (1999) Phylogeny of Hypselodoris (Nudibranchia: Chromodorididae) with a review of the monophyletic clade of Indo-Pacific species, including descriptions of twelve new species. Zoological Journal of the Linnean Society 125: 1–125.
- 14. Gosliner TM (2001) Aposematic coloration and mimicry in opisthobranch mollusks: new phylogenetic and experimental data. Bollettino Malacologico 37: 163–170.
- 15. Cortesi F, Cheney KL (2010) Conspicuousness is correlated with toxicity in marine opisthobranchs. Journal of Evolutionary Biology 23: 1509–1518.
- 16. Wilson NG (2002) Egg masses of chromodorid nudibranchs (Mollusca: Gastropoda: Opisthobranchia). Malacologia 44(2): 289–306.
- 17. Wilson NG, Lee MSY (2005) Molecular Phylogeny of Chromodoris (Mollusca: Nudibranchia) and the identification of a planar spawning clade. Molecular Phylogenetics and Evolution 36: 722–727.
- 18. Coelho R, Calado G (2010) Spawn and early development of NE Atlantic species of Hypselodoris (Gastropoda: Opisthobranchia). Iberus 28(2): 63–72.
- 19. Lee WL, Devaney DM, Emerson WK, Ferris VR, Hart CW Jr, et al. (1978) Resources in invertebrate systematics. American Zoologist 18: 167–185.
- 20. Knowlton N (1993) Sibling species in the sea. Annual Review of Ecology and Systematics 24: 189–216.
- 21. Knowlton N, Jackson JBC (1994) New taxonomy and niche partitioning on coral reefs: jack of all trades or master of some? Trends in Ecology and Evolution 9: 7–9.
- 22. Committee on Biological Diversity in Marine Systems, National Research Council (1995) Understanding Marine Biodiversity. National Academy of Sciences.
- 23. Valdés Á (2002) A phylogenetic analysis and systematic revision of the cryptobranch dorids (Mollusca, Nudibranchia, Anthobranchia). Zoological Journal of the Linnean Society 136: 535–636.
- 24. Brooks DR, McLennan DA (1991) Phylogeny, Ecology and Behavior: A research Program in Comparative Biology. 434 p. The University of Chicago Press. Chicago, Illinois.
- 25. Williams DM, Kociolek JP (2007) Pursuit of a natural classification of diatoms: History, monophyly and the rejection of paraphyletic taxale. European Journal of Phycology 42(3): 313–319.
- 26. Rudman WB (1984) The Chromodorididae (Opisthobranchia: Mollusca) of the Indo-West Pacific: a review of the genera. Zoological Journal of the Linnean Society 81: 115–273.
- 27. Turner LM, Wilson NG (2008) Polyphyly across oceans: a molecular phylogeny of the Chromodorididae (Mollusca, Nudibranchia). Zoologica Scripta 37: 23–42.
- 28. Johnson RF (2011) Breaking family ties: taxon sampling and molecular phylogeny of chromodorid nudibranchs (Mollusca, Gastropoda). Zoologica Scripta 40(2): 137–157.
- 29. Gosliner TM (1996) Phylogeny of Ceratosoma (Nudibranchia: Chromorididae), with descriptions of two new species. Proceedings of the California Academy of Sciences 49(3): 115–126.
- 30. Johnson RF, Valdés Á (2001) The Hypselodoris infucata colour group (Mollusca, Nudibranchia, Chromodorididae), with remarks on the genus Brachychlanis Ehrenberg, 1831. Journal of Natural History 35: 1371–1398.
- 31. Alejandrino A, Valdés Á (2006) Phylogeny and biogeography of the Atlantic and eastern Pacific Hypselodoris Stimpson, 1855 (Nudibranchia, Chromodorididae) with the description of a new species from the Caribbean Sea. Journal of Molluscan Studies 72: 189–198.
- 32. Thollesson M (1999) Phylogenetic analysis of dorid nudibranchs (Gastropoda: Doridacea) using the mitochondrial 16S rRNA gene. Journal of Molluscan Studies 65: 335–353.
- 33. Thollesson M (2000) Increasing fidelity in parsimony analysis of dorid nudibranchs by differential weighting, or a tale of two genes. Molecular Phylogenetics and Evolution 16: 161–172.
- 34. Wollscheid E, Wägele H (1999) Initial results on the molecular phylogeny of the Nudibranchia (Gastropoda, Opisthobranchia) based on 18s rDNA data. Molecular Phylogenetics and Evolution 13: 215–226.
- 35. Wollscheid-Lengeling E, Boore J, Brown W, Wägele H (2001) The phylogeny of Nudibranchia (Opisthobranchia, Gastropoda, Mollusca) reconstructed by three molecular markers. Organisms, Diversity and Evolution 1: 241–256.
- 36. Grande C, Templado J, Cervera JL, Zardoya R (2004a) Molecular phylogeny of Euthyneura (Mollusca: Gastopoda). Molecular Biology and Evolution 21: 303–313.
- 37. Grande C, Templado J, Cervera JL, Zardoya R (2004b) Phylogenetic relationships among Opisthobranchia (Mollusca: Gastropoda) based on mitochondrial cx1, trnV, and rrnL genes. Molecular Phylogenetics and Evolution 33: 378–388.
- 38. Gosliner TM, Draheim R (1996) Indo-Pacific opisthobranch gastropod biogeography: how do we know what we don't know. American Malacological Bulletin 12(1/2): 37–43.
- 39. Debelius H (1998) Nudibranchs and Sea Slugs-Indo-Pacific Field Guide. IKAN-Unterwasserarchiv. Frankfurt.
- 40. Gosliner TM, Behrens DW, Valdés Á (2008) Indo Pacific Nudibranchs and Sea Slugs. 425 p. Sea Challengers/California Academy of Sciences.
- 41. Baker RJ (1994) Some thoughts on conservation, biodiversity, museums, molecular characters, systematics and basic research. Journal of Mammalogy 75(2): 277–287.
- 42. Medina M, Walsh PJ (2000a) Molecular Systematics of the Order Anaspidea based on mitochondrial DNA sequence (12s, 16s and COI). Molecular Phylogenetics and Evolution 15: 41–58.
- 43. Medina M, Walsh PJ (2000b) Comparison of four Mendelian loci of the California sea hare (Aplysia californica) from populations of the coast of California and the Sea of Cortez. Marine Biotechnology 2: 449–455.
- 44. Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology 3: 294–299.
- 45. Palumbi S, Martin A, Romano S, McMillan WO, Stice L, et al. (1996) The simple fool's guide to PCR, Version 2.0. Hawaii: Department of Zoology and Kewalo Marine Laboratory, University of Hawaii.
- 46. Maddison DR, Maddison WP (2005) MacClade version 4.08. Sinauer Asssociates.
- 47. Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32(5): 1792–1797.
- 48. Swofford DL (2002) PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4. Sinauer Associates, Sunderland, Massachusetts.
- 49. Nylander JAA (2004) MrModeltest v2. Program distributed by the author. Evolutionary Biology Centre, Uppsala University. Available: http://www.abc.se/~nylander/. Accessed 2012 February 28.
- 50. Huelsenbeck JP, Ronquist F (2001) MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755.
- 51. Ronquist A, Huelsenbeck JP (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572–1574.
- 52. Nylander JAA, Ronquist F, Huelsenbeck JP, Nieves-Aldrey JL (2004) Bayesian phylogenetic analysis of combined data. Systematic Biology 53: 47–67.
- 53. Rambaut A, Drummond AJ (2007) Tracer v1.4, Available from Tracer website Available: http://tree.bio.ed.ac.uk/software/tracer/. Accessed 2012 Feb 28.
- 54. Felsenstein J (2001) The troubled growth of statistical phylogenetics. Systematic Biology 50: 465–467.
- 55. Felsenstein J (2004) A digression on history and philosophy. In: Felsenstein J, editor. Inferring Phylogenies. pp. 123–146. Sinauer Associates, Sunderland, MA.
- 56. Franz NM (2005) On the lack of good scientific reasons for the growing phylogeny classification gap. Cladistics 21: 495–500.
- 57. International Commisson of Zoological Nomenclature (2000) International Code of Zoological Nomenclature. Fourth Edition. The International Trust for Zoological Nomenclature c/o The Natural History Museum London, United Kingdom. Available: http://www.nhm.ac.uk/hosted-sites/iczn/code/. Accessed 2012 Feb 28.
- 58. Bouchet P, Rocroi JP, Frýda J, Hausdorf B, Ponder W, et al. (2005) Classification and nomenclator of gastropod families. Malacologia 47(1–2): 1–397.
- 59. International Society for Phylogenetic Nomenclature (2010) The Phylocode. Version 4c. Available: http://www.ohio.edu/phylocode/. Accessed 2012 Feb 28.
- 60. Dayrat B, Gosliner TM (2005) Species names and metaphyly: a case study in Discodorididae (Mollusca, Gastropoda, Euthyneura, Nudibranchia, Doridina). Zoologica Scripta 34(2): 199–224.
- 61. Gray JE (1857) Guide to the systematic distribution of Mollusca in the British Musuem. Part 1. London Taylor and Francis.
- 62. Thiele J (1931) Hanbuch der Systematischen Weichtierkinde: 431. Jena: Fischer.
- 63. Risbec J (1928) Contribution à l'étude des nudibranchs Néo Calédoniens. Faune des colonies Francaises 2: 328.
- 64. Odhner NHJ (1934) The Nudibranchiata. British Antarctic (“Terra Nova”) Expedition, 1910. British Museum of Natural History Report Zoology 7(5): 229–310, pls.1–3 P 251.
- 65. Bertsch H (1977) The Chromodoridinae nudibranchs from the Pacific coast of America-Part I. Investigative methods and supra-specific taxonomy. Veliger 20: 107–118.
- 66. Odhner NHJ (1968) Systematique, In A. Franc, Sous-classe des Opisthobranches, Traite de Zoologie (P. Grasse Ed.) 5: 834–888.
- 67. Bergh LSR (1898) Die Opisthobranchiata der Sammlung Plate. Fauna Chilensis. Zoologische Jahrbucher Suppl. 4: 481–582.
- 68. Ortea JA (1988) Molluscos Opisthobranquios del Archipelago de Cabo Verde: Chromodorididae. Publicações Ocasionais da Sociedade Portuguesa de Malacologia 11: 1–16.
- 69. Valdés Á, Afonso CML (2003) Synonymy of Cadlina burnayi Ortea, 1988, from the Cape Verde Islands (Opisthobranchia: Chromodorididae) with Tyrinna evelinae (Marcus, 1958). The Veliger 46(2): 176–179.
- 70. Muniain C, Valdés Á, Ortea J (1996) Redescription of Tyrinna nobilis Bergh, 1898 (Opisthobranchia: Chromodorididae) from Patagonia, Argentina. Journal of Molluscan Studies 62(3): 265–273.
- 71. Schrödel M, Millen SV (2001) Revision of the nudibranch genus Tyrinna Bergh, 1898 (Doridoidea: Chromodorididae). Journal of Natural History 35: 1143–1171.
- 72. Baba K (1937) Opisthobranchia of Japan 2. Journal of the Department of Agriculture Kyushu Imperial University 5(7): 289–344.
- 73. Pruvot-Fol A (1931) Notes de systematique sur le opisthobranches. Bulletin du Museum National d'Histoire Naturelle, Paris (2) 3(3): 309–310.
- 74. Basedow H, Hedley C (1905) South Australian nudibranchs and an enumeration of the known Australian species. Transactions and Proceedings of the Royal Society of South Australia 29: 134–160.
- 75. Rudman WB (1986) The Chromodorididae (Opisthobranchia: Mollusca) of the Indo-West Pacific: Noumea flava colour group. Zoological Journal of Linnean Society 86: 377–404.
- 76. Ehrenberg CG (1831) Symbolae physicae seu incones et descriptions animalium evertebratorum sepositis insectis quae ex itinere per Agricam Borealem et Asiam Occidentalem. Decas 1 Mollusca.
- 77. Rudman WB (1986) The Chromodorididae (Opisthobranchia: Mollusca) of the Indo-West Pacific: The genus Glossodoris Ehrenberg ( = Casella, H.A. Adams). Zoological Journal of Linnean Society 86: 101–184.
- 78. Pease WH (1866) Remarks on Nudibranchiata inhabiting the Pacific Islands, with descriptions of two new genera. American Journal of Conchology 2(3): 204–208.
- 79. Odhner NHJ (1934) The Nudibranchiata. British Antarctic (“Terra Nova”) Expedition, 1910. British Museum of Natural History Report Zoology 7(5): 229–310, pls.1–3 P 251.
- 80. Cheeseman TF (1881) On some new species of nudibranchiate Mollusca. Transactions and Proceedings of the New Zealand Institute 13: 222–224.
- 81. d'Orbigny A (1839) Mollusques, echinoderms, foraminiferes et polypiers, recueillis aux Iles Canaries par Mm. Webb et Berthelot et decrits par Alcide D'Obrigny. Mollusques-1 2(2): 1–117, pls 1–76.
- 82. Cuvier G (1804) Memoire sur le genre Doris. Annales du Museum National d'Histoire, Paris 4: 447–473.
- 83. Herrmannsen AN (1847) Indicus generum malacozoorum primordial. Vol. 1. Cassellis.
- 84. Adams H, Adams A (1858) The Genera of the Recent Mollusca, vol. 2. London: Van Voorst.
- 85. Marcus Ev, Marcus Er (1967) American opisthobranch mollusks. Studies in Tropical Oceanography 6: 1–256, pl. 1.
- 86. Marcus E (1971) On some euthyneuran gastropods from the Indian and Pacific Oceans. Proceedings of the Malacological Society, London 39: 355–369.
- 87. Alder J, Hancock A (1855) Monograph of the British Nudibranciate Mollusca. Appendix. London. Ray Society.
- 88. Quoy JR, Gaimard JP (1832) Voyage de decouvertes de l'Astrolabe pendant les annees 1826–1829 sous le commendement de M.J. Dumont d' Urville Zoologie 2: 1–686.
- 89. Rudman WB (1987) The Chromodorididae (Opisthobranchia: Mollusca) of the Indo-West Pacific: Chromodoris epicuria, C. aureopurpurea, C. annulata, C. coi and Risbecia tryoni colour groups. Zoological Journal of the Linnean Society 90: 305–407.
- 90. Bergh LSR (1891) Die cryptobranchiaten Dorididen. Zoologische Jahrbucher 6: 103–144.
- 91. Bergh LSR (1874) Neue nacktschnecken der Südsee, malacologische Untersuchungen. 2. Journal des Museum Godeffroy 2(6): 91–116, pls.1–4.
- 92. Bergh LSR (1875) Neue nacktschnecken der Südsee, malacologische Untersuchungen. 3. Journal des Museum Godeffroy 3(8): 53–100 (185–232), pls.7–11.
- 93. Valdés Á, Gosliner TM (1999) Reassessment of the systematic position of the systematic status of Miamira Bergh 1875 and Orodoris Bergh, 1875 (Nudibranchia: Chromodorididae) in light of phylogenetic analysis. Journal of Molluscan Studies 65: 33–45.
- 94. Adams A, Reeve L (1850) Mollusca, Part 3. The Zoology of the Voyage of H.M.S. Samarang during the years 1843–46. Reeve, Benham & Reeve: London.
- 95. Bertsch H (1976) A new species of Chromodoris (Opisthobranchia: Nudibranchia) from tropical West America. The Veliger 19: 156–158.
- 96. Bergh LSR (1878) Malacologische Untersuchungen 13. In: Semper C, editor. Reisen im Archipel Philippinen. pp. 495–546. Wissenschaftliche Resultate. Vol. 2, no. 2.
- 97. Burn R (1961) A new doridid nudibranch from Torquay, Victoria. The Veliger 4: 55–56, pl.15.
- 98. Stimpson W (1855) Descriptions of some new marine Invertebrates. Proceedings of the Academy of Natural Sciences, Philadelphia, 7(10): 385–395.
- 99. Rudman WB (1983) The Chromodorididae (Opisthobranchia: Mollusca) of the Indo-West Pacific: Chromodoris splendida, C. aspersa and Hypselodoris placida colour groups. Zoological Journal of the Linnean Society 78: 105–173.
- 100. Gosliner TM, Ghiselin MT (1984) Parallel evolution in opishthobranch gastropods and its implications for phylogenetic methodology. Systematic Zoology 33: 255–274.
- 101. Gosliner TM, Johnson S (1994) Review of the genus Hallaxa (Nudibranchia: Actinocyclidae) with descriptions of nine new species. The Veliger 37(2): 155–191.
- 102. Ortea J, Valdés Á, Garcia-Gomez JC (1996) Review of the Atlantic species of the family Chromodorididae (Mollusca: Nudibanchia) of the blue chromatic group. Avicennia Supplement.
- 103. MacFarland FM (1966) Studies of opisthobranchiate mollusks of the Pacific coast of North America Memoirs of the California. Academy of Sciences 6: 1–546.
- 104. Kirkendale LA, Meyer CP (2004) Phylogeography of the Patelloida profunda group (Gastropoda: Lottidae): diversification in a dispersal-driven marine system. Molecular Ecology 13: 2749–2762.
- 105. Latiolais JM, Taylor MS, Roy K, Hellberg ME (2006) A molecular phylogenetic analysis of strombid gastropod morphological diversity. Molecular Phylogenetics and Evolution 41: 436–444.
- 106. Frey MA, Vermeij GJ (2008) Molecular phylogenies and historical biogeography of a circumtropical group of gastropods (genus: Nerita): implications for regional diversity patterns on the marine tropics. Molecular Phylogenetics and Evolution 48: 1067–1086.
- 107. Reid DG, Dyal P, Lozouet P, Glaubrecht M, Williams ST (2008) Mudwhelks and mangroves: the evolutionary history of an ecological association (Gastropoda: Potamididae). Molecular Phylogenetics and Evolution 47: 680–699.
- 108. Malaquias MAE, Reid DG (2009) Tethyan vicariance, relictualism and speciation: evidence from a global molecular phylogeny of the opisthobranch genus Bulla. Journal of Biogeography 36: 1760–1777.