Revision of salamandrid taxonomy


Site Contributor
Nov 8, 2002
Reaction score
Manchester, England
United Kingdom
Display Name
Chris Michaels
Has anyone seen this paper and has it been accepted? Dubois and Raffaelli make quite a few changes to the Salamandridae; some additional ranks (subfamilies (Salamandrinae, Pleurodelinae, Salamandrininae), tribes (e.g. Molgini and Pleurodelini within the Pleurodelinae)etc) are added, and some splits are made (e.g. Hypselotriton is reinstated for Chinese Cynops) as well as some other changes.

Dubois and Raffaelli (2009) A new ergotaxonomy of the family Salamandridae Alytes, 2009, 26 (1-4): 1-85.

Free online here: - file attached.


(PS you have to wade through a huge discussion of taxonomic theory, so skip to half way down page 29 for the changes to the Salamandridae)


  • A new ergotaxonomy of the family Salamandridae, Dubois and Raffaelli.pdf
    1.6 MB · Views: 2,639
Last edited:
I posted it last year, and it has been a source of taxonomical fun for me ever since i first heard of it.
As far as i know (which is not saying much) the changes haven´t been accepted, but i personally agree with the vast majority. I love the revised taxonomy for Cynops, and the re-classification of the european newts, it´s beautifully done and to me it makes the utmost sense.
Has anyone seen this paper and has it been accepted? Dubois and Raffaelli make quite a few changes to the Salamandridae; some additional ranks (subfamilies (Salamandrinae, Pleurodelinae, Salamandrininae), tribes (e.g. Molgini and Pleurodelini within the Pleurodelinae)etc) are added, and some splits are made (e.g. Hypselotriton is reinstated for Chinese Cynops) as well as some other changes.

Dubois and Raffaelli (2009) A new ergotaxonomy of the family Salamandridae Alytes, 2009, 26 (1-4): 1-85.

Free online here: - file attached.


(PS you have to wade through a huge discussion of taxonomic theory, so skip to half way down page 29 for the changes to the Salamandridae)

Wow Froggy. Interesting and at same time insane... Freaks me out the hybrid stuff...

Pleurodeles waltl x Tylototriton verrucosus
Triturus x Lissotriton
Mesotriton x Lissotriton
Mesotriton x Triturus
Lissotriton x Ommatotriton

I wonder witch species they use to acomplish this... and hope they surely were destroyed afterwards. InterGenera hybrids might be more than odd... I would like to see some pictures of this.:kill:

Jean sent me the paper right after it was published. I agree with many of the conclusions, though not all.

As to whether it's "accepted", I think that reflects a common misconception about taxonomy - very few things are "official". Acceptance is up to the reader to determine, although part of that is determining whether the methods used and conclusions made were actually appropriate. For instance, regarding Molgini as a tribe rather than Molginae as a subfamily, is pretty much in agreement with everything published in a very long time. However, it's simply a matter of preference. Adherents of phylocode tend to split everything in two. That means that they will only accept two subfamilies normally, and any others will become families or tribes within subfamilies. At higher levels, they are quite happy to use unranked categories paired with families or groups of families. Where Salamandridae is concerned, it makes little difference, except to make clear that newts are closer to Pleurodeles than either is to Salamandra. I prefer to recognize newts as a subfamily, as overall they differ distinctly from the Pleurodeles line, and have a similar distribution [which indicates to me that they evolved in parallel into different niches].

I think that many of the subgenera proposed are unnecessary, though potentially valid. All they do in this paper is identify which of two species a third is most closely related to. That's the same principle used in phylocode, but a rather superfluous number of names if applied across the animal kingdom [or even amphibians] as a whole. Imagine - two subgenera for every three species! Another reason for many of the names is basically for Dubois to enforce his personal nomenclatural doctrine. They [Dubois really] argue that names should be drastically shortened for ease of use. By naming every possible grouping according to his own standards, the rest of the world is forced to follow his standard on the principle of "first valid proposed name". I personally find scientific names to be quite informative BECAUSE they often use full root words, suffixes, and prefixes which have specific meanings in the languages of origin. Shortening them arbritrarily obscures these root meanings. His arguments are also based on a purely eurocentric point of view - although Linnaean nomenclature is Latin in rules and European in origin, it is intended to be universal in application. Many names now have origins [and authors] in languages which are quite comfortable with longer words. I have no problem with Protohynobius puxiongensis. I have no problem with Lyciasalamandra. I DO have a problem with Algandra, which completely obscures BOTH root words, Algeria and Salamandra. I would be comfortable with Algirandra, as it at least makes one of the roots clear.

Also because of the limited analysis involved and the very constrained groupings (one or two species), it is likely to prove that some of the subgenera are invalid. Further studies are bound to show that the type species of one subgenus is actually closer to a member of a different subgenus, rendering one of the names a useless synonym.

For such an extensive treatment, I have to admit to being somewhat appalled at the data used to reach the conclusions. Ease of captive care? Since when is that a trait which can definitively be assigned to the organism? Ease of captive care is a *human* trait, in that we have to figure out what's required. If an organism doesn't do well in our care, I would argue that it's OUR fault for not getting things right, not the organism's fault for being difficult, and not a reflection of species relationships.

On the other hand, some things were overdue. Salamandra longirostris was clearly a distinct species needing elevation, the African Salamandra were clearly several distinct species, and Laotriton was definitely distinct from Paramesotriton. The systematics they use for other Paramesotriton is clearly premature, as the evidence I have looked at shows that some species they placed arbitrarily in subgenus Paramesotriton are actually related to Allomesotriton (P.caudopunctatus). That also renders the distinction of Allomesotriton questionable, as the distinctive physical traits may only apply to one species in the clade! I don't recall then applying a plethora of subgenera to Paramesotriton, even though they felt the need to split the three species of Notophthalmus and the four of Taricha. The need for splitting of Cynops is long overdue. It was suggested decades ago that Cynops should be three genera, corresponding to Cynops, Hypselotriton (Hypselotriton) and Hypselotriton (Pingia) as applied by Dubois and Raffaelli. I have argued elsewhere as to the usage of Pingia, as it's STILL a synonym of Pachytriton. That leaves the orientalis group as Hypselotriton until such time as a new name is proposed.

Overall, it's an interesting read from a nomenclatural philosophy point of view. It's also quite useful from the perspective of genera and species recognized. The subgenera are of mixed value, and the subfamilies and tribes are a bit of a case of personal preference. Other recent works which add phylocode to the mix (Frost et al) recognize only two subfamilies, but they didn't include Salamandrina, which has been possibly the most confusing and uncertain member of the family. They place Salamandrina as a basal Pleurodelinae. Another study placed it close to Tylototriton, and others have placed it closer to Salamandra, Triturus, or basal within the family! The only consistent conclusion is that it is distinctive. I prefer to recognize Salamandrinae, Salamandrininae, Pleurodelinae, and Molginae. As long as Salamandrina isn't buried close to another genus, this will be valid in most situations, and it also recognizes morphological, behavioral, and geographic trends within the family. As long as a name has been validly proposed, anyone is free to interpret the data and adopt the names as they feel best. Within the bounds of priority and monophyly [oldest names apply; a name should be applied to a group which includes ALL related species]

mods, if this is a duplicate post, please delete the earlier unrevised version.
Thanks for your comments, Frogeyes. By 'accepted' I meant, has this paper had an impact on the majority of taxonomists and other workers using the taxonomy, rather than being 'agreed' by some non-existent central authority on salamander taxonomy.

I guess that depends on your perspective. I think that very little of it is likely to be put to use, such as most of the subgenera.

Laotriton is in widespread use, and that's a name proposed in this paper.
Full species of Salamandra are in more widespread use, though I'm not sure how much of that can be attributed to Dubois and Raffaelli.
I don't think most of the names will have an impact. Most of the subgenera are unnecessary, and subgenera aren't used much in the first place. The subfamilies and tribes aren't significant revisions.
Hypselotriton was overdue for revalidation, as was the Pingia grouping [notwithstanding that the name isn't valid for these species], but these are not yet in common usage.

I think the main purpose of the paper was to provide a platform for one more of Dubois' nomenclatural arguments, in this case the shortening of names. I doubt that will have much impact either. Dubois seems to have nothing better to do than create enough new technical terms to make any taxonomic paper unreadable except to those who've taken several university courses on terminology.

The parts I personally find most important are Yaotriton and Laotriton named, Hypselotriton revalidated, species of Salamandra raised, and perhaps recognition of Algiandra [though not all subgenera of Salamandra].
I´m being tempted to start using Hypselotriton...xD
I just might, Janusz :p
It ´s not easy to eliminate Cynops orientalis from one´s brain...being such a charismatic and well known name, but certainly the taxonomical change is necessary. I just wonder if as FrogEyes said, a new name could be proposed rendering Hypselotriton invalid.

I kind of like it now, has grown on me...Hypselotriton.....nice.

I agree the shortening of the names is quite ridiculous. You can infer quite a lot of information from certain scientific names, it´s a pitty to loose that just to gain in simplicity. Algandra is a slap on the face, it sounds just awful, and i agree, you can barely distinguish its origin unless you know what it refers to.

I personally quite like the subgenera, if only for the information it provides on phyllogeny, which is something that i find very interesting indeed. Granted, they have very little further use and are quite messy.
Hypselotriton, thus far, seems to be a rather secure taxon name. It was established based on Cynops wolterstorffi and is most appropriately applied to the clade which includes that species plus C.cyanurus and C.chenggongensis.

This clade is most closely related to the C.orientalis species group, which also includes C.orphicus and C.fudingensis [plus potentially a species from northern Zhejiang]. Consequently, the C.orientalis group must also be transferred to Hypselotriton in order to prevent Cynops being used for two unrelated groups. This is a temporary measure, since it was long ago suggested that the differences between the three groups of "Cynops" was equal to the differences between full genera of other newts. The only reasons these are included in Hypselotriton for now are these: they are most closely related to Hypselotriton; there is no other available name at the moment; they are not closely related to Cynops pyrrhogaster.

Hypselotriton should be used, but one should expect it to change to a new name at some point, where the small Chinese newts are concerned.

I've discussed the application of the names Pingia and Pingia granulosa in a Pachytriton brevipes thread. Although redescribed as a purported member of the Hypselotriton orientalis group, the name Pingia granulosa belongs to a Pachytriton. Juvenile Pachytriton strongly resemble Hypselotriton; which is no shock, given that some Paramesotriton strongly resemble Pachytriton, and Hypselotriton is closer to Pachytriton and Paramesotriton than it is to Cynops. Whether the new specimens assigned to Pingia are truly a new species of Hypselotriton remains to be seen, but identical animals in Anhui are juvenile Pachytriton. As a result, there may or may not be a new species in the H.orientalis group, but the name Pingia granulosa does not apply to it.
General chit-chat
Help Users
  • No one is chatting at the moment.
    sera: @Clareclare, +2