Andinoacara

[cichla]

Musilová, Z, O. Říčan and J. Novák (online): Phylogeny of the Neotropical cichlid fish tribe Cichlasomatini (Teleostei: Cichlidae) based on morphological and molecular data, with the description of a new genus. J.Zool. Syst. Evol. R.

ABSTRACT

Phylogenetic relationships among cichlasomatine cichlids were studied using an extensive taxon sampling and both morphological and molecular data sets. A new genus, Andinoacara n. gen. with six species (A. pulcher-rivulatus group of previous authors) from trans-andean South America and NW cis-andean South America, is described based on results of phylogenetic and diagnosability analyses and tests of alternative topologies Our results demonstrate that cichlasomatine cichlid diversity is divided into five principal lineages composed of eleven genera and three suprageneric clades: the [(Bujurquina, Tahuantinsuyoa), (Andinoacara) (BAT) clade; the (Cleithracara, (Nannacara, Ivanacara)] clade (NIC) plus Laetacara and 'Aequidens' hoehnei; and the (Aequidens, Cichlasoma) clade, where Aequidens is paraphyletic to Cichlasoma. Two former Aequidens species are additionally transferred into Krobia (K. potaroensis, K. paloemeuensis). 'Aequidens' hoehnei probably represents a unique evolutionary lineage and would thus qualify for a separate generic status. Molecular data are yet not available for this species and its generic status requires further study. Relationships between the three suprageneric clades and between Acaronia and Krobia could not be convincingly resolved with our data set of two mitochondrial (16S and cyt b) and two nuclear (S7 and RAG1) molecular markers and 96 morphological characters.

[Philippe Burnel]

Complete publication available here : http://www3.interscience.wiley.com/cgi- ... 0/PDFSTART

Thanks Ingo


Philippe

[Willem Heijns]

Interesting conclusion of the study at hand is that the name Aequidens is about to disappear from the scene. Musilová et al found out that the natural group containing the Aequidens species as currently understood also contains the whole of Cichlasoma. Instead of defining four new genera for the species to be removed from Aequidens because of the paraphyly of this genus, the authors suggest the two genera should be grouped into one (monophyletic) genus. Fore this genus the oldest available name would be Cichlasoma, rendering Aequidens to a synonym.

[Tachymarptis]

I am not sure that it is so interesting, Willem. It is a dogmatic perversion of cladistic methodology to absolutely want to suppress paraphyletic taxa; this leads to "taxonomic black holes" in large classification sections (see for example the questionable shrinking of Xenotilapia and related taxa by Takahashi). We ought to stop being dogmatic on this point and to admit that in linnean nomenclature, it is impossible to do without plesions (= paraphyletic taxa), unless taxinomic ranks (including genera and species) are suppressed (and even...).

[Willem Heijns]

In my opinion it is interesting, Tachymarptis (btw: why would Cichlid Room Experts want to hide behind a nickname?). To me it is quite logical to try and define taxa as natural groups. After all it reflects history and relationships in nature. Saying that in Linnean nomenclature it is impossible to do without paraphyletic taxa can be just as dogmatic.

In this case the close relationship between Aequidens and Cichlasoma at least calls for a discussion about wether to keep these two genera apart or join them into one. What do you think?

[Juan Artigas]

I am not a fan of merging genera on the basis of monophyly, after all, genera is supposed to tell us that a plesiomorphic character has had so much success in evolution of a taxa that it has been inherited to its derived species, or has produced a big change in an organism (monotopic genera). So, I agree with "Tachymarptis" that cladistic classification should recognize paraphyletic trees and derived genera (after all man made rather subjective decisions).

[Bas Pels]

I think taxonomists should try to keep groupes monophyletic - that is, untill we have information evolution results in paraphyletic groupes (which I don't expect, but that aside)

However, genetics will never be able to group any number of species, or even define a species.

Some species are much more diverse than others. Therefore, a XX% similarity in genes will not be able to define any species or group. That is where wisdom takes over from chemistry and mathamatics

Wisdom is needed to draw lines in the genetic tree, to define groupes and exclude others

Back to Aequidens and Cichlasoma, if one would combine them, one would need to have good arguments: Why are the found differences so unimportant that they are to be seen as 1 group?

Whether Aequidens disappears into synonimity or not does not matter to me, I rather have a taxonomity which is according to the latest insights

[Tachymarptis]

In my opinion it is interesting, Tachymarptis (btw: why would Cichlid Room Experts want to hide behind a nickname?).

I am not hiding, Willem, I have reasons to do so; you (or anyone) only have to send me a personal message, or better, an e-mail, you will be be welcome. Anyway, this is not important.

To me it is quite logical to try and define taxa as natural groups.

I fully agree with this since I began to be fond of animal evolution during my childhood (since the beginning of the 70's). Let me say that I was already calling birds "flying dinosaurs" well before this allegory became "in the mood" (in fact, dinosaurs shoud better be defined as walking birds, this was their first identification in the beginning of the XIXth century, when the first tracks were found, before their skeletons were found and before the name "dinosaurs" was proposed). I was first enthusiast when the first classification upheavals came, but I found later that there were limitations to what we can do as "natural groups". These limitations are related to our historical classification in species, genera and so on, which was formalized by Linnaeus.

After all it reflects history and relationships in nature.

Let me put ahead an important fact which is often completely bypassed by current taxonomists: classifications were already attempting to reflect phylogeny, probably since Linnaeus himself. The trend, with more or less formalization, was to reflect life evolution and to avoid polyphyletic groupings. Cladistics have not invented phylogeny, it only is a good methodology to solve phylogenetic problems -and it is also an approximation. Of course, this speeded up the classifications improvements, as well as the arrival of new techniques (biomoleculars). But cladistic methods are also approximations.

Saying that in Linnean nomenclature it is impossible to do without paraphyletic taxa can be just as dogmatic.

I do not state that gratuitously. You will find as many examples as you want. Species give birth to other species, so unless they vanish completely, species are paraphyletic taxa. The reasoning is same for genera, which are, or are to become paraphyletic taxa if they do not disappear. For higher taxa, you could possibly (with great difficulties) keep them only as clades, but all ranks become obsolete, taxa become simple matryochka dolls with very few significance for use by human mind (hence Juan-Miguel's remark, which is shared by many people).

By the way, in case of monophyletic taxa, I use "clades" and "grades" (or "plesions"). In the litteral sense, paraphyletic taxa are monophyletic, i.e. from a single ancestor, even if they do not encompass all their descendants.

In this case the close relationship between Aequidens and Cichlasoma at least calls for a discussion about wether to keep these two genera apart or join them into one. What do you think?

Maybe it is of interest to merge Cichlasoma and Aequidens, but for the reasons given above, it must not be only for coping with this "paraphyly rule". Applying this rule without discernment is, in my opinion, dogmatic (sorry to insist). This clearly is the case with the "Xenotilapia black hole", which upsets most of us. But for Aequidens and Cichlasoma, why not? But with other arguments than the only "paraphyletic rule".

Sincerely,

Patrick.

[Willem Heijns]

Thanks for your reply Patrick.

An interesting note on the dinosaur/bird/reptile problem. To me that is a good example of a classification without knowledge of phylogeny. If Linneaus would have known the history of these animals, birds would have been reptiles or dinosaurs. Part of the problem is that many of the species studied are extinct. So we will probably never know if the ancestral species of all birds was actually a bird or just another dinosaur (assuming birds arose from the dinosaurs).

Closer to home (i.e. cichlids) I think the dynamic view to evolution/speciation is a difficult concept. You are right by saying that in a historical perspective every species is or will become paraphyletic. If a small population gets (reproductively) isolated from its conspecifics and becomes the founder of a new species, than the ancestral species automatically becomes paraphyletic. In this case the ancestral species probably doesn't change (much). But if the isolated population is larger (maybe even as large as half of the original population) it gets more complicated. Which of the two resulting populations is the ancestor? What if both populations change (considerably)?

Another important factor is speed. Speciation takes time. I believe that most of the phylogenetic studies are studies of a situation (a moment in time) reflecting the current relationships between species (and higher taxa). That is probably why none of these studies (as far as I know) designates extant species as ancestors or descendants of other extant species. All studied taxa are terminal taxa and all we can do is make statements about the relationships between those terminal taxa and their hypothesized ancestors.

Even closer to home (i.e. Aequidens/Cichlasoma) the issue is not about splitting or joining two related taxa. If Aequidens and Cichlasoma were two sister taxa, I can only agree with Juan Miguel when he says that merging them on the basis of monophyly is not mandatory. I have made the same comment on the Hypsophrys/Neetroplus issue. But as current knowledge indicates all Cichlasoma species are placed well within the phylogeny of Aequidens (as we know it today). Seen from the dynamic view Aequidens has at some point in time become paraphyletic. So what?
On the other hand I believe the phylogeny of the species at hand is exactly the same as when Aequidens (1894) and Cichlasoma (1839) were described. The point is that nobody knew at the time. Had Eigenmann and Bray and/or Swainson known the true relationships, their genera would have been defined differently, containing different species. What I am trying to say is that the more we learn, the better our taxonomic definitions become. That is why generic assignments change.

[cichla]

Dear all,
there are (at least) two different schools of systematics: the Evolutionary Systamatic (defended by Ernst Mayr, the ''Darwin of the 20th century'') and the ''Phylogenetic Systematic'' (developed by Willi Hennig), which is later transformed into cladistics. The traditional view of the Evolutionary Systematics allow to keep paraphyletic units (if these are not polyphyletic). The strict cladism accept only monophyletic assemblages. Hence, it is more a case of philosophy. However, nowadays the cladistics and PhyloCode take place and ''our beloved'' Linnean nomenclature (genus and species names) is threatened.

Aequidens: In this case, all Aequidens could be included into Cichlasoma (this would not that new. To quote Ahl (1924): „.. dass sich wohl in Kürze die Notwendigkeit ergeben wird, die Gattungen Acara (=Aequidens) und Cichlasoma zu vereinigen...“ [... that probably in the near future the need arise, to unit the genera Acara (= Aequidens) and Cichlasoma...]. As a result we would have a monophyletic assemblage. However, following this procedure nobody could create a new name. This is why I am sure, that in near future a new genus name will be appear (at least diadema-gruop; see Smith et al. 2008).
Greetings, IS

BTW: It is not sure that birds are descents from dinosaurs (see Mayr 2001: What Evolution is).

[Tachymarptis]

Another important factor is speed. Speciation takes time. I believe that most of the phylogenetic studies are studies of a situation (a moment in time) reflecting the current relationships between species (and higher taxa). That is probably why none of these studies (as far as I know) designates extant species as ancestors or descendants of other extant species. All studied taxa are terminal taxa and all we can do is make statements about the relationships between those terminal taxa and their hypothesized ancestors.

Yes, evolution speeds or freezes depending on circumstances. Small isolated populations evolve faster, thus giving new species while the parental species is still unchanged (see for example cichlids of lake Nabugabo, originating from Victoria haplochromines). Depending on the selection pressure, evolution may accelerate or remain quiescent during long periods. This was called "punctuated equilibrium" by Stephen Gould and Niles Eldredge.

And you are not right when supposing that cladistic studies are used mainly for present-day situations. Paleontologists are greatly influenced by cladists and follow the "no-paraphyly" dogma, despite it is impossible to apply when naming ancestral taxa. This constantly surprises me, and leads to uninterrupted contortions in classifications with few added informations, but it is.

Anyway, cladists are now becoming aware of the logical incoherences of their reasoning. I do not agree with you, Cichla, this is not only a matter of philosophy, it is a question of logic. That's why they are looking towards the phylocode which, as I stated, fits better than the Linnean classification. But even if they succeed in establishing a more "mathematical" nomenclature (this is by no mean evident, as evolution is not an exact, but rather a descriptive science), this one would not have any usefulness for the uninitiated. And unless both nomenclatures evolve together, which anyway would mean giving up the cladist's inapplicable constraints in the linnean one, this would be a great failure. People would then use classifications unrelated to the actual knowledge, or even go back to old, declining nomenclature.

PS: About birds, their origin from dinosaurs or other archosaurs such as crocodilians has long been debated, but now, evidences are growing more and more each year, and now, specialists which still doubt that birds come from theropod dinosaurs are becoming increasingly rare.

[Livio Leoni]

An interesting post about a match between Linnean and cladistic classification.
PS: Anyway about BAD (Birds are dinosaurs) and BAND (birds are not dinosaurs) hypotheses I can suggest you these posts of this Italian blog.
http://theropoda.blogspot.com/2009/05/b ... vadis.html
http://theropoda.blogspot.com/2009/05/b ... rte-i.html
(You can translate these posts with google translator and I think you can understand quite well.)
At the moment seems BAD hypotheses is the most parsimonious hypothesis.

Livio

[Livio Leoni]

I don't remember the theory very well, but none of the resulting populations is the ancestor.

Livio

Thanks for your reply Patrick.
But if the isolated population is larger (maybe even as large as half of the original population) it gets more complicated. Which of the two resulting populations is the ancestor? What if both populations change (considerably)?

[Rico Morgenstern]

It is a dogmatic perversion of cladistic methodology to absolutely want to suppress paraphyletic taxa

You will find as many examples as you want. Species give birth to other species, so unless they vanish completely, species are paraphyletic taxa. The reasoning is same for genera, which are, or are to become paraphyletic taxa if they do not disappear. For higher taxa, you could possibly (with great difficulties) keep them only as clades, but all ranks become obsolete, taxa become simple matryochka dolls with very few significance for use by human mind (hence Juan-Miguel's remark, which is shared by many people).

There's a good example in Cichlids: the species flock of Lake Barombi Mbo, which consists of eleven endemic species in five genera, four of which are also endemic. It is in all all probability derived from a single, extant subspecies - Sarotherodon galilaeus galilaeus, which is therefore a paraphyletic taxon. To avoid paraphyly, one had to synonymize all the Barombi Mbo species with that one subspecies, which would make absolutely no sense to evereone who knows the species. There are some examples of "DNA" papers, where it seems doubtful if the authors know at all the creatures behind character matrixes, gene sequences and so on.
Cladistics can only be a tool, a method to investigate relationships by calculating statistical probabilities, but the results are not the absolute truth and should therefore be treated which caution. It all depends from the suitability of chosen characters, taxon sampling and choice of outgroups, and of course of the knowledge of the species.

Regarding the Aequidens - Cichlasoma problem, the author's decision to defer it to a more detailed study is very wise. So we should too wait and see, and meanwhile leave the genera as they are.

The description of a new genus Andinoacara was certainly overdue, one can only hope that a species level revision will follow soon, there is not only the A. rivulatus problem, the limits between species of 'Blue Acaras' also have yet to be established.

All the best
Rico