Template talk:Laurasiatheria Cladogram

As per comments I made on Talk:Laurasiatheria regarding a 2011 study by Zhou et al, Pegasoferae appears to lack support and should be deprecated. Following Zhou, the revised Laurasiatheria Cladogram should be something like this:

Laurasiatheria	Eulipotyphla    Ferungulata     Chiroptera    Fereuungulata       Ferae     Pholidota  Carnivora    Euungulata     Perissodactyla      Cetartiodactyla

Note these pairs of terms are roughly interchangable:

• Ferungulata, Scrotifera
• Fereuungulata, Cetferungulata
• Euungulata, Cetungulata

-- Limulus (talk) 11:55, 3 October 2011 (UTC)

Also, while Euungulata is favored, the data does not "fully support" it (compare with "trees reconstructed by BI and ML methods for the 1,608-gene data set fully support (100%/1.00/100%) a basal position for Eulipotyphla and a more apical position for Chiroptera"), so to indicate that this is not fully resolved, perhaps:

Laurasiatheria	Eulipotyphla    Ferungulata     Chiroptera    Fereuungulata       Ferae     Pholidota  Carnivora  Perissodactyla      Cetartiodactyla

Also of interest, Table S4 from the supporting online material for Impacts of the Cretaceous Terrestrial Revolution and KPg Extinction on Mammal Diversification (2011) 'fully supports' a monophyletic Eulipotyphla, "Ostentoria" (= Ferae) and "Variamana" (= Ferungulata).

-- Limulus (talk) 15:27, 3 October 2011 (UTC)

There have been a lot of different phylogenetic hypotheses on Laurasiatheria interrelationships; I don't think it's a good idea to jump to using the exact topology of any new paper. Cf. WT:MAMMAL#Laurasiathere taxonomy. Ucucha (talk) 21:07, 3 October 2011 (UTC)

Compared to the old version, what I wanted to emphasize is that Pegasoferae is not showing up in newer genetic studies. In fact, this newest one is specifically saying that "Pegasoferae (Perissodactyla + Carnivora + Pholidota + Chiroptera) does not appear to be a natural group." I think that there is enough published material to support the second ("not fully resolved") cladogram I posted above with Chiroptera as the outgroup to Ferae+ungulates. -- Limulus (talk) 22:38, 3 October 2011 (UTC)

Your version is definitely an improvement to the previous version; I'd have to review the literature more to be convinced that Fereuungulata (a name designed to be unpronounceable, I suppose) is well enough supported. Ucucha (talk) 22:56, 3 October 2011 (UTC)

Starting with PubMed results for Fereuungulata, which lists six papers, the origin of the term is:

• Waddell et al (1999) "The elephant sequence allows confident rejection of the older taxon Ferungulata (Simpson, 1945). In its place, the new taxa Scrotifera and Fereuungulata are defined." (for more details see Page 50 (PDF page 20)) They reference:

• Pumo et al (1998) using a bat mitochondrial genome "All analyses placed bats as the sister group of carnivores, perissodactyls, artiodactyls, and cetaceans (e.g., 100% bootstrap value with both maximum parsimony and neighbor joining). The data strongly support a new hypothesis about the origin of bats, specifically a bat/ferungulate grouping."

• and Waddell et al (1999b) "The support for bats plus fereuungulates forming a clade was near 100%, while the position of bats sister to the fereuungulates (i.e., consistent with the monophyly of Fereuungulata) was supported near the 90% level, but with some uncertainty."

• Nikaido et al (2000) using a different bat mt genome "Maximum-likelihood analysis of 12 concatenated mitochondrial proteins from 36 mammals strongly suggested the monophyly of the order Chiroptera and its close relationship to Fereuungulata (Carnivora + Perissodactyla + Cetartiodactyla)."

• Cao et al (2000) A Chiroptera/Eulipotyphla (i.e. bat/mole) clade and a close relationship of this clade to Fereuungulata (Carnivora+Perissodactyla+Cetartiodactyla) were reconfirmed with high statistical significance. However, a support for a monophyly of Fereuungulata relative to the Chiroptera/Eulipotyphla clade was fragile, and we suggest that the three branchings among Carnivora, Perissodactyla, Cetartiodactyla and Chiroptera/Eulipotyphla occurred successively in a short time period, estimated to be approximately 77 Myr BP.

• Waddell et al (2001) "In addition, there is good evidence for a clade of all placental mammals except Xenarthra and Afrotheria (=Boreotheria) and for the previously recognised clades Laurasiatheria, Scrotifera, Fereuungulata, Ferae, Afrotheria, Euarchonta, Glires, and Eulipotyphla." Note Fig. 3 "Arrows indicate local rearrangements that might yet be correct." (specifically the two: Ferae+Perissodactyla and Eulipotyphla+Chiroptera but in either case Chiroptera is an outgroup to FEuU)

• then Nishihara et al (2006) introduced Pegasoferae but as noted in its article it was not supported by subsequent studies:

• Springer et al (2007) (see Fig . 1) supported a clade of (Pholidota+Carnivora)+(Perissodactyla+Cetartiodactyla), with bats as an outgroup. The posterior probabilities are all nicely 100 *except* for the node linking the ungulates, so if we collapse that, we get the "not fully resolved" cladogram :) I can't access Matthee et al (2007) here but maybe you can to see if that is a good ref for Fereuungulata.

• Kitazoe et al (2007) show two trees in Fig. 1 both of which are the first cladogram I posted w/o Pholidota.

• and finally, Zhou et al (2011) which specifically says "Pegasoferae (Perissodactyla + Carnivora + Pholidota + Chiroptera) does not appear to be a natural group." and "Our reconstructions resolve the interordinal relationships within Laurasiatheria and corroborate the clades Scrotifera, Fereuungulata, and Cetartiodactyla."

So IMHO it is well supported. Also, I totally agree that the name is a tongue-twister BTW... I guess they can't all be cool and fun like Whippomorpha though ;) -- Limulus (talk) 10:34, 4 October 2011 (UTC)

The Matthee et al. (2007) paper actually has Chiroptera sister to Carnivora + Pholidota (with Cetartiodactyla, Perissodactyla, and Soricidae as successive outgroups to that); it has very uneven sampling, though (concentrated in bovids, bats, and leporids). doi:10.1093/molbev/msn104 (Prasad et al. 2008) has Chiroptera sister to Cetartiodactyla + Perissodactyla, but with only 20% bootstrap support. doi:10.1186/1471-2148-8-162 (Hallström and Janke 2008), which has many genes but few species, couldn't resolve interordinal relationships within Scrotifera. Asher and Helgen (2010; http://www.biomedcentral.com/1471-2148/10/102) in a review article don't use Fereuungulata in their composite consensus tree. Zhou et al. (2011) say that basal branches in Scrotifera are quite short, which probably accounts for the difficulty in resolving phylogenies. Ucucha (talk) 13:12, 4 October 2011 (UTC)

Thank you for checking Matthee; the Pegasoferae article was incorrect then (fixed now). Fig. 1 of Asher & Helgen has Ferae as the outgroup to the rest of Scrotifera which I haven't seen before; I do note that Table 2 mentions both Pegasoferae and Fereuungulata as being not depicted in Fig. 1 but having "potential support" so that unfortunately doesn't clear up anything (a cladogram with just Eulipotyphla and Scrotifera isn't overly helpful IMHO :). I think relying on Zhou (for 'fully supported' clades; even they couldn't fully resolve FEuU) until a yet better paper is published would be the best idea for now. -- Limulus (talk) 13:57, 4 October 2011 (UTC)

I just had a chance to read through the Asher & Helgen paper; Tables 1 and 2 might form the basis of an article and certainly would help in creating redirect pages. Their sources for Fig. 1 are specifically:

• Prasad, et al. (2008) which admits: "the arrangement of orders within Laurasiatheria appears to be difficult to resolve even with large amounts of sequence data and reasonably large numbers of species represented. We further found that the relative arrangement of Laurasiatherian orders was highly sensitive to alignment guide tree artifacts, though not in a predictable way. Using the coding plus conserved noncoding sequence matrix, we performed SH tests with the 5 most supported Laurasiatherian trees from the literature; none could be excluded with high confidence, and this likely is due, in part, to the short branches separating Laurasiatherian orders. Perhaps with increased taxon sampling, this problem will be more tractable." The Fig. 1 tree fully supports a Eulipotyphla/Scrotifera split, but the other nodes are very weak (to the point where I would just collapse them for Wikipedia). Fig. 2 tree is much closer to what we're used to seeing (basically the 'not fully supported' tree if you collapse the weakest nodes and keep the 98% Chiroptera node)

• Murphy, et al (2007) in Fig. 6 appears to derive its datapoints from articles from 2003 and 2004 and has a weird arrangement of Scrotifera that resolves cetart. but forms a chir+peri clade.

In short, I do not trust the Laurasiatheria tree in Asher & Helgen :) -- Limulus (talk) 21:10, 4 October 2011 (UTC)

etc after Carnivora listing

I advocate that this is appropriate in distinguishing the incompleteness of this listing in comparison with the completeness of all the other listings. For example, in Chiroptera subclades are not listed, but are obviously all covered under "bats and flying foxes". In Artiodactyla, given that "whales" is an obvious stand-in for cetaceans, all extant clades are covered (review the definition of "ruminants"). In contrast, the Carnivora listing is missing many major clades; all musteloideans, most viverroideans (given that "hyenas" are not an obvious stand-in for the latter group), Asiatic linsangs and palm civets. There is thus a distinct dichotomy in completeness of listings, and it's useful to indicate that. WolfmanSF (talk) 20:30, 12 March 2017 (UTC)

I take it that "this" means "the use of 'etc.'" when labelling a clade. If so, yes, it's highly appropriate, and indeed wise. The current labels are rather long (remember people often read articles on small mobile screens) and would probably benefit from shortening, with 'etc.' commonplace. Chiswick Chap (talk) 20:42, 12 March 2017 (UTC)

What I meant was after Carnivora listing only (because all the other listings are pretty complete at present), rather than shortening the listings, but anything is up for discussion. WolfmanSF (talk) 22:57, 12 March 2017 (UTC)

I'd suggest that more than one "etc." would be needed for the present structure, but the list effect is not good and is causing edit-warring as people start to feel uncomfortable: I think they're basically right, something is wrong. Here's why. Cladograms such as this have to fit into several articles, and in recent weeks the width on screen has gone up by 20%. We can't possibly list all clades within each group, nor illustrate more than one animal for each leaf node, however tempting it may seem to include a marine mammal as a surprising exception - that should be dealt with in the appropriate articles. If we want to show that sort of diversity within a cladogram, it should be by adding further nodes so we get a richer tree, but still with one label (and one image) per leaf node. Anything else looks and feels cluttered, because people are trying to use words and images to indicate branching, which a cladogram can handle naturally by actually branching. In other words, we're going wrong by using lists of English subclade names, and lists of images within single leaf nodes. Chiswick Chap (talk) 11:47, 13 March 2017 (UTC)

The listings seem reasonable except for the Carnivora and Cetartiodactyla. The Carnivora list in particular shows a bias towards the larger carnivores, with the listing of cats, hyaenas, dogs, bears and seals. None of the smaller carnivores – mustelids, mongooses, civets and the like – get mentioned, despite their diversity. The listing of ruminants also seems out of place as we hardly talk about ruminants in common speech. It would be better to select one or more (cow, deer?) as an example(s), but this just gets to the problem of the expanding list. As the lists are not inclusive, I favour including etc or an equivalent. What would be nice is to have collapsible/expanding cladograms, so lower subdivisions could be easily displayed, but this is difficult with the current system and wikipedia restrictions. Jts1882 (talk) 17:15, 13 March 2017 (UTC)

We can certainly add a little more detail to the cladogram's branching, however. Chiswick Chap (talk) 18:05, 13 March 2017 (UTC)

The problem there is that some of the most interesting (or surprising) branches are embedded quite deeply. I'm thinking of the seals and whales. It would be more appropriate in some articles than others, so perhaps not ideal for the template. This discussion has given me a couple of ideas on interactive possibilities for those willing to add gadgets. Either a simple hide/show option, something like the Hovercard, or something more dedicated. I'll look at a few things tomorrow. Jts1882 (talk) 20:53, 13 March 2017 (UTC)

If the template is even slightly getting in the way, we should abandon it and use slightly more specialised cladograms in each article. After all, if you have a tree with, say, five branches (each one quite twiggy) then in article A you want a cladogram showing A in detail and B, C, D, E in outline (or perhaps a little more detail on B); for the article on B you want more detail on B, with a bit on A and C, and an outline of D and E. The implication is that a template is unlikely to be the right answer in any of the five cases. I've found in fish and insect cladograms, for instance, that I usually want to specialise for each group, even though they of course overlap. Chiswick Chap (talk) 02:00, 14 March 2017 (UTC)

At the moment all lists are basically inclusive except for Carnivora, where we have focused on the more charismatic groups. An inclusive list for Carnivora using common terms is impossible. "Ruminants" is reasonably common, but less so than the other terms; that was a compromise to make possible an inclusive Artiodactyla listing. Note that a general cladogram like this is often used in conjunction with a more specialized cladogram for the focus of an article (i.e., I see no reason to abandon this one). I honestly disagree that the balance struck between branching and listing is wrong here; it's just different than some might be used to.

If we must narrow the template, we could reduce the Carnivora listing to "cats, dogs, seals, etc" and the Artiodactyl listing to "pigs, deer, whales, etc" and/or eliminate the 2nd photo for each. But do we need to? It fits easily on the width a a standard screen and isn't causing problems in that regard anywhere. I can't speak to mobile usage, but mobile users must be used to dealing with lots of compromises. WolfmanSF (talk) 04:16, 14 March 2017 (UTC)

Artiodactyla versus Cetartiodactyla

I recently changed the template to use Cetartiodactyla, which on my understanding has become the accepted name for the clade containing the artiodactyls with the cetaceans. This was reverted by @WolfmanSF: and this in turn reverted by @Manudouz:. It seems a good time for a discussion.

I can list a host of articles supporting the use of Cetartiodactyla. However, there was the proposal by Spaulding et al (2009) to use Artiodactyla as the name with precedence. They make a case that I am sympathetic with, and also make a good case against Whippomorpha. My understanding was that it hasn't been widely adopted and that Cetartiodactyla is still used more widely. For instance, the comprehensive study of mammalian relationships by Meredith et al (2013) used Cetartiodactyla. Is there an authoritative source that we can use to make the decision on which is the best choice now? Jts1882 (talk) 14:05, 14 March 2017 (UTC)

This is a recurrent discussion when more recent phylogenies do not fit the traditional systematics. There are here two options: (i) either keep the traditional taxon names to account for precedence and to avoid introducing too many new names ; (ii) or introduce new names corresponding to monophyletic groups, and abandon the older names because they refer to para-/poly-phyletic groups.

The first option involves redefining the content of traditional groups: this might be ambiguous, and also confusing for general readers.

The second option introduces new names which might become more and more used. I would prefer this option because it is unambiguous, and redirect (generalist / specialist) readers to the discovery of new taxonomic terms.

For example, extant "reptiles" have been changed to archosaurs (birds + crocodiles), archelosaurs (archosaurs + turtles), and sauropsids (archelosaurs + lepidosaurians). Manudouz (talk) 15:09, 14 March 2017 (UTC)

"Cetacea" now tends to be viewed as an infraorder within Artiodactyla. "Artiodactyla" defined in this way is a monophyletic clade and never needed to be renamed. We use "Artiodactyla" in the taxonomy of both the Artiodactyla and Cetacea articles. The higher level mammal taxonomic articles I have read recently {e.g. #1, #2) use "Artiodactyla". There's no particular reason we have to be 10 years behind the curve in this shift in usage, as Wikipedia so often is. WolfmanSF (talk) 16:30, 14 March 2017 (UTC)

That sounds sensible, but if we are going to do this, it would help enormously if Even-toed ungulate (where A. redirects) were updated, with citation, to say clearly that A. is the name to use, not Cetartiodactyla which is what the article uses right now. Chiswick Chap (talk) 16:38, 14 March 2017 (UTC)

I think that is just one view and I'm far from convinced that it is universal. Some people use Artiodactyla and others Cetartiodactyla, something the article ultimately has to reflect. I just wrote something in the Even-toed_ungulate talk page which summaries the dilemma in my eyes. Jts1882 (talk) 17:05, 14 March 2017 (UTC)

You are convincing me more and more that a template, which necessarily embodies one view of the phylogeny, is a poor solution here. Each article can and should contain a cited discussion of the evidence that applies to it, and editors can choose to tailor their own cladograms to suit the individual taxon, both (as mentioned above) by cutting down detail on less relevant branches of the tree, and by naming taxa appropriately to the evidence found. The one-size-fits-all approach is not bringing consensus - if anything it seems further off now than when we started, and we are finding more and more reasons to disagree in terms of cladistics, layout, amount of detail, number illustrations and so on. Chiswick Chap (talk) 10:19, 15 March 2017 (UTC)

Editors are free to edit the template version in articles when it is important (use {{Subst:templatename}} and then make changes). The template still serves a purpose as a tool people can use, either directly or as a starting point. The cladogram in the template just needs to be the most generally useful version. To me that means it should be as up to date as possible and not too cluttered. Jts1882 (talk) 11:44, 15 March 2017 (UTC)