Trout and other salmonid taxonomies are still in a chaotic state and in many respects have advanced little since the 19th Century. Salmonids are renowned for their phenotypic plasticity expressed under different environmental conditions. This high plasticity in many morphological characters and life histories is such that almost any population will be found to differ from other populations especially if only a few populations are compared. Yet such characters are the basis of many species descriptions. Some claim to be following the Evolutionary Species Concept (ESC) of Simpson (1951), where “An evolutionary species is a lineage evolving separately from others and with its own unitary role and tendencies”. Evidence for the ESC is provided by morphological differences that are adaptive in nature (my emphasis), i.e., by definition have a genetic basis (Simpson, 1961). Yet many simply assume that the morphological differences that they use have a genetic and adaptive basis without further investigation even though heritability may be extremely low or absent. In that respect their approach is purely phenetic.
Since most conservation legislation is species-based accurate taxonomy is key to conservation of salmonid biodiversity. Bad taxonomy can kill by failing to recognise a population as a distinct taxon and thus it does not receive the conservation attention it requires. On the other, it can result in wasted conservation resources if the taxon is based on purely environmentally-induced differences and is simply part of a more widespread species of lesser concern. Some 51 species of Salmo trouts are currently recognised in FishBase and recent publications, including several in recent years. Most trout species have been classified on colouration, spotting pattern, occurrence of parr marks in adults, dentition, scale counts, and body measurements. In many, but not all situations, these characters are subject to environmental modulation with the effects of phenotypic plasticity and adaptation being difficult to disentangle. Body measurements are, in some cases, converted to ratios of standard length, but this approach has long been regarded as inappropriate due to allometric growth. Often insufficient specimens and populations are examined to give a true picture of intra- and inter-population variability.
An important criterion in taxonomy is that the characters used to define a species can be used to identify individuals to that species with ≥ 99% of individuals being correctly assigned (Mayr, 1963), either using molecular approaches or genetically based life history and morphological differences. Etheridge et al (2012) found that the power of supposedly diagnostic morphological characters to identify individuals of three putative Coregonus species was low (27%) due to the species descriptions being based on a few specimens, and as a result of phenotypic plasticity.
Given that a reference sequence is available for brown trout and that the determination of full genomic sequences is now relatively straightforward, is there any reason why a DNA sequence in an appropriate depository cannot be the name-bearing type sequence for a species? Linked to the type nuclear sequence should be DNA specimens, which can be used for further study. Once isolated, it can be stored nearly indefinitely. DNA can be easily shared for secure, multi-site curation. Since it takes up little space and can be stored at room temp there is no reason why all national museums should not be involved in such curation. Mitochondrial DNA sequences, while much easier to obtain, are problematic due to the potential for horizontal transfer and the linkage of genes. There are several examples of incongruence between nuclear and mtDNA. Use of only part of the nuclear genome could also be potentially problematic due to differentiation between some closely related trout being present in localised genomic ‘islands’. Sufficient DNA sequences to represent intra-specific variability would be required. Clearly international collaboration would be required to cover the entire Salmo trout range and a meaningful number of specimens. Do others consider this a potential way forward and what are the possible difficulties involved? Or is the real question whether conservation legislation should be species-based in the first place but instead be focused on populations, or groups of populations, as in North America using Evolutionarily Significant Units or Designatable Units?
Concerning the original question - most importantly, at this time, the idea of a DNA sequence serving as a type specimen is of course in complete violation of the Zoological Code, and I am not sure it would make much sense to change this. That being said, zoological collections and taxonomic specialists are increasingly including DNA or preserved tissue adequate for genetic analysis in their collections, and there is a growing number of young taxonomists who strive to incorporate molecular genetic analysis into their skill set. Further, most national Bar-Coding projects (as is the case in Austria), demand parrallel DNA storage with all specimens. This is a great advance, that will slowly contribute to perhaps resolving at least some of the problems outlined above - it is a shame however, if, for example, many of these newly described Salmo taxa do not have DNA or adequately preserved tissue for future genetic analysis. This creates a stumbling block if one wishes to go back and incorporate these specimens into a larger meta analysis with the hopes of streamlining the current chaos.
I completely agree with you! I am also convinced that only extensive international collaboration would help resolve the intricate taxonomy of brown trout (Salmo trutta complex). On the other hand, I am absolutely convinced that conservation legislation (as well as management strategies) should be based not so much on species - too unstable and too coarse-grained - but instead be focused on populations, evolutionary units "par excellence", than in freshwater fish often correspond to the river basin or sub-basin in which they evolved.
Vincenzo Caputo Barucchi Thank you for your contribution to the discussion.
I agree with your point about protecting populations, or groups of populations, as distinct conservation units (ESUs, DUs or whatever). However, irrespective of whether a conservation unit or species is used it is still necessary to consider the appropriate approach to determining that the unit / species is distinct. Conservation units or Evolutionary Species should only be described in terms of adaptive genetic differences, that is, not the result of genetic drift.For the most part morphological characters are of little value for this purpose and generally represent phenotypic plasticity. Thus, whether one is describing a conservation unit or a species, the same considerations apply. However, a species may comprise two or more units thus indicating their monophyly. It is the real “shame of European ichthyology” that species are still being described using phenotypically plastic characters and without adequate genetic data. Preserved museum specimens are now redundant and we need to move forward to DNA collections. One of the problems of conservation units is the lack of recognition in conservation legislation in most countries. It will certainly be a difficult and time consuming task to get conservation units accepted by bodies such as IUCN and statutory conservation bodies. The alternative is not to designate the organism but the river or lake where it occurs thus protecting the entire community. Such geographical site designations are already widely in the use in the EU and the UK.
The problem is not the choice of species concept. You mention the Evolutionary species concept sensu Simpson who was a paleontologist. I use the species concept but according to Wiley and Mayden (2000) for who the time is important and there is the notion of "young species". In my opinion, we cannot delineate a species with only molecular data which is a tool among others. This data should be corroborated with at least morphological characters. So whatever the genetic distance of the evolutionary lineages, the importance is that they are diagnosable with morphological data.
The main problem with brown trouts lineages is that molecular data do not often corroborated with morphology because of polymorphism. There are several examples in Salmo spp taxonomy. Moreover taxonomists and phylogenetists disagree about their species delineations.
However, the International Code of Zoological Nomenclature has some strict rules which must be respected. You can consider evolutionary lineages as distinct conservation units if and only if they belong to the same species. And for that, there must be written in a publication "Salmo X is junior synonym of Salmo trutta Linnaeus, 1758". Without this mention, the name (e.g. Salmo cettii, Salmo labrax, Salmo multipunctata, etc..) is still valid.
There is nothing in the Zoological Code that defines what a species is, the code merely provides the rules for nomenclature. The Code explicitly states that biologists and not the Code describe species. Further, and to the surprise of many how have not bothered to read it, subspecific nomenclature is in full concordance with the Zoological Code. Finally, legislation alone defines what type of biological unit is provided protection. The United States, Canada, and I believe Australia as well if I am not mistaken, provide protection under their laws for biological units below the level of species - in theory, any legislative body in the world could pass a law to protect a population, or metapopulation, or lineage, however defined. There is no inherent connection between the aims of conservation legistlation (generically speaking) and taxonomic status. I understand that many countries protect only species, but that is by choice, and is not bound by any scientific theory. Thus, I do not see the desire to protect biodiversity necessitating an increasing number of species. Further, I generally support most if not all of Andy's comments about Salmo, for years I have been informally saying the same thing - many of these new species are described with substandard data sets that take neither the phenotypic nor genetic data of the genus into proper consideration, among other problems.
Concerning the original question - most importantly, at this time, the idea of a DNA sequence serving as a type specimen is of course in complete violation of the Zoological Code, and I am not sure it would make much sense to change this. That being said, zoological collections and taxonomic specialists are increasingly including DNA or preserved tissue adequate for genetic analysis in their collections, and there is a growing number of young taxonomists who strive to incorporate molecular genetic analysis into their skill set. Further, most national Bar-Coding projects (as is the case in Austria), demand parrallel DNA storage with all specimens. This is a great advance, that will slowly contribute to perhaps resolving at least some of the problems outlined above - it is a shame however, if, for example, many of these newly described Salmo taxa do not have DNA or adequately preserved tissue for future genetic analysis. This creates a stumbling block if one wishes to go back and incorporate these specimens into a larger meta analysis with the hopes of streamlining the current chaos.
Gaël Denys
I can appreciate the value in some cases of integrating information from genetics, morphology, ecology, behaviour, etc but I don’t see why molecular data should only be accepted if corroborated with morphology. There are, however, many types of molecular data from a few markers such as microsatellites to full genomic sequences. I am referring only to the latter. ‘Evolutionary’ species (in the broadest sense), as well as conservation units (CU), are based on adaptive differences - genetic difference by itself is not sufficient. Microsatellites can reveal reproductively isolated populations, but generally not adaptation, and can be used to discriminate management units in fisheries, but not species or CUs. Genomic sequences can demonstrate distinctness as well as adaptation even where morphological differences are absent. Although the paper does not have a taxonomic focus, a recent preprint by Saha et al (2021 - DOI: 10.22541/au.161035029.98155231/v1) demonstrates substantial sequence differences (FST=0.13) between two populations of brown trout from Lakes Bunnersjöarna in Sweden. These sympatric populations were originally identified with allozymes and do not show any obvious morphological differences. Putative adaptive divergence was identified in a number of relevant genes.
Steven Joseph Weiss
Sorry, I got that confused. The sequence would be part of the obligatory publication. However, a DNA specimen could perhaps act as a type - ICZN 72.5.1. “an animal, or any part of an animal”.
One problem that I perceive with current DNA collections is that they are often in the collections of individual laboratories rather than in a central museum with the obligation and facilities to make aliquots available to researchers. However, clearly not the case in Austria from what you say.
There are many national bar coding projects that are functioning identical to Austria's, and there are a number of museums that are also archiving tissue and DNA, and making aliqots available for research, I am burrowing from a few right now. There is a lot of positive movement in that direction, but this does not involve what has been going on with "Salmo" lately.
Hi Andrew,
Because I am interested in the topic "genetic but no morphological divergence" in teleost fishes just a short note.
I think your example concerning sympatric brown trout populations, genetically divergent but not morphologically, is not an appropriate one.
Although Saha et al. 2021, DOI: 10.22541/au.161035029.98155231/v1 mention in the Conclusions: “Our work demonstrates that populations from the same habitat may have large genome-wide divergence without obvious morphological distinction ….”, I could not find morphological data, investigations or what ever morphological in this study. This study is purely genetically. In the introduction I found a citation which refers to the same sympatric brown trout populations (Andersson et al. 2017. DOI 10.1007/s10592-017-0972-4). These authors say that these brown trouts cannot be “discriminated by the naked eye” but that the populations show “minor tendencies for morphological divergence” and cite for this Andersson et al. 2016. DOI: 10.1111/eff.12308. These authors say: “We find small significant differences for body shape, body size …”. Significant differences in morphological traits – this does not sound for no morphological differences.
To my experience, fish populations which are genetically divergent but not morphologically, are very, very rare, if there are some at all. Mostly standard morphological traits get investigated in these studies. I have the feeling that some times “no morphological differences” just means “no morphological differences found”.
All the best,
Harald
Harald Ahnelt
Many thanks for your interesting comments. As far as I am aware no detailed study of morphometrics and meristics has been published for the Bunnersjöarna populations. However, in other sympatric situations the populations are clearly discriminated by the naked eye as, for example, the Lough Melvin sympatric species that we have studied in some detail (see my RG page for papers). I appreciate that in many cases where sympatric populations have been identified on genetic grounds morphological differences are subsequently found (e.g. Fišer et al 2018 review, albeit mainly on amphipods).
My main reason for citing the Saha et al preprint was to show that resequencing of the Salmo genome is now underway (in several laboratories to my knowledge), and that classification based on full genomic sequences, and potential adaptations, is now feasible.
Uwe Zajonz
Many thanks for your valuable comments. As I acknowledged in my response to Steve Weiss, it would have to be a DNA specimen as a type - ICZN 72.5.1. “an animal, or any part of an animal”. The sequence would be part of the obligatory publication.
As you say, there are many approaches to the conservation of biodiversity. However, in my limited experience I have found species arguments more likely to be accepted than population genetics ones. Undoubtedly this varies with the background of the individuals concerned and is likely to vary considerably among jurisdictions.
My main issue is that salmonid species are still being described on phenotypically plastic characters of low heritability and with inappropriate and incorrect data analyses. Where molecular data are added these are often very limited and not appropriate for demonstrating adaptations. Undoubtedly preserved museum species are of value, more so in some groups than in others. Appropriately taken photographs (avoiding photographic distortion) are also of value for geometric morphometrics. As full genomic sequencing becomes even easier it is likely to be applicable to all species, not just commercially important ones. It is therefore vital that DNA specimens are required now in all cases alongside traditional ones in museum collections. It would be interesting to have a museum curator’s perspective on this.
Regarding the morpological vs. molecular conundrum, I don't think any revision of the ICZN is needed.
The value of a deposited specimen mostly resides in the wealth of undescribed information that its biological structure potentially contains. We simply don't know what further information can be gleaned from a preserved specimen.
Therefore, no improvement could come from replacing a part of this structure (e.g., morphology) with another (e.g., genomic information). Tissue samples for molecular analyses and genomic sequences can be simply added to the lot including the morphological specimens (that could also be preserved with different methods; e.g., both formalin- and ethanol-fixed), as well as any other documentation (e.g., photos of freshly dead or anaesthetised specimens, morphometric measurements before fixation, X-rays, ecological notes, environmental data on the habitat). This is already common practice in several large museums, though it is obviously left to the individual researcher depositing the lot. Encouraging this practice is certainly commendable.
On the other hand, I think that a considerable improvement may come from preserving and depositing morphological specimens of populations that genomic analyses have demonstrated to be reliable evolutionary units. The deposition of such material, both morphological and molecular, could certainly be a strong basis for defining a taxonomic and/or conservation unit, something the conservation biology of salmonids urgently needs.
Hi Andy,
thank you very much for the interesting discussion you started. Below, the link for a very new paper, in my opinion, helpful for the arguments issues in the above debating.
Article Human impacts on global freshwater fish biodiversity
- Badges
- Science topic