Mammal phylogeny employing fossil and extant taxa

Current hypotheses of relationships among mammals
may cause one to develop a large cartoon question mark over one’s cranium. Traditional paleontologists still cannot recover the closest sisters to bats and they continue to suggest that hippos are the closest sisters to known pre-whales with legs.

very few fossil taxa are employed in these analyses (send them along if you know of any). DNA studies don’t completely match morphological studies and they ignore fossil taxa. Wikipedia sums up the current thinking dividing paleontologists from molecular biologists.

Molecular Biologists recover this tree (highly simplified):

  1. Atlantogenata =
    Xenartha (sloth + anteater + armadillo)
    Afrotheria (golden mole + elephant shrew + tenrec + aardvark + hyrax + elephant + sea cow) Afroinsectiphilia (golden mole + elephant shrew + tenrec),
  2. Boreoeutheria (Boreotheria) =
    Euarchontoglires (rodent +  rabbit + tree shrew + flying lemur + primate + Plesiadapis)
    Laurasiatheria (hedge hog + mole + shrew + bat + pangolin + carnivores + odd-toed ungulates + even toed ungulates + whales)

Traditional paleontologists recover this tree (highly simplified):

  1. Xenarthra (sloth + anteater + armadillo)
  2. Afrotheria (hyrax + elephant + sea cow)
  3. Boreoeutheria (Boreotheria) =
    Euarchontoglires (rodent +  rabbit + tree shrew + elephant shrews + flying lemur + primate + Plesiadapis)
    Laurasiatheria (hedge hog + mole + shrew + bat + pangolin + carnivores + odd-toed ungulates & aardvark + even toed ungulates & whales)

So there is broad agreement between the two camps.
The problem comes when one tries to replicate the experiment (Fig. 1). Adding mammal taxa to the large reptile tree delivers a different hypothesis of relationships. It preserves some traditional relationships and recovers some new ones. If you’re not sure why this keeps happening here, it happens on the professional level, too. Some prior workers employ suprageneric taxa. Others do not employ fossil taxa. I also wonder if there is substantial convergence in the dental traits of included taxa. Teeth vary greatly within the Mammalia.

The tree (built taxon by taxon)
presented here (Fig. 1) seems to make more sense (more sister taxa look similar to one another) with a basal split between large arboreal omnivores (that ultimately produced carnivores) and small arboreal omnivores (that ultimately produced large and small herbivores and some insectivores).

Figure 1. The family tree (cladogram) of mammal interrelationships. Here the basal division is between carnivores and insectivores + herbivores.

Figure 1. The family tree (cladogram) of mammal interrelationships. Here the basal division is between slightly larger arboreal omnivores and slightly smaller arboreal omnivores. Of course, evolution kicks into high gear once  the dinosaurs are gone.

Similarities with traditional trees:

  1. Outgroups include cynodonts, monotremes and marsupials in that order.
  2. Eomaia is a basal placental (eutherian) despite retaining prepubic bones.
  3. Primates nest together and with a flying lemur
  4. Bats nest with carnivores
  5. Civets nest together
  6. Rodents nest with rabbits and one tree shrew
  7. Xenarthans nest together
  8. Ungulates nest together with Phenacodus (55 mya) at their base
  9. Elephant nests with hyrax

Differences with traditional trees:

  1. There are no geography-based divisions here, only morphology
  2. The basic division is between arboreal civet-like omnivores and arboreal rodent-like omnivores.
  3. Pangolin nests with primates and Plesiadapis does not
  4. Tree shrews are not a single clade. Tupaia nests with rodents, rabbits and Plesiadapis. Ptilocercus nests with flying lemur.
  5. Bats nest with a specific carnivore: Chriacus.
  6. Whales nest with tenrecs and pre-tenrecs, not ungulates
  7. Aardvark nests with armadillo derived from Pantolambda, which has a carnivore-like appearance).
  8. Ungulates nest with elephant + hyrax all derived from a sister to Phenacodus.

In the large reptile tree, all sister taxa share a long list of traits
and look similar to one another. You can’t say that about competing hypotheses which don’t always include fossil taxa. The mammals subset of the large reptile tree (725 taxa) has grown so far without tree topology changes. Taxa just drop in between existing nodes.

Very few tooth characters are used here.
And none distinguish one post-canine tooth from another other than [absent, blunt, sharp, multi cusp and multi cusp with constricted base] and another trait notes where the posterior maxillary tooth erupts relative to the orbit.

Wikipedia suggests that:
“Paleontologists naturally insist that fossil evidence must take priority over deductions from samples of the DNA of modern animals. More surprisingly, these new family trees have been criticised by other molecular phylogeneticists, sometimes quite harshly.”

We’ve come a long way…
Simpson 1945 was able to list several subclasses and orders, but was unable to show interrelationships. Novacek 1992 discussed several problems right before computer-assisted phylogenetic analysis came along.

Suggestions for pertinent mammal taxa that need to be added to the LRT?
Please, send them along.

Novacek MJ 1992. Mammalian phylogeny: shaking the tree. Nature Review Article. 356:121-125.
Simpson GG 1945. The principles of classification and a classification of mammals. Bulletin of the AMNH Bulletin of the American Museum of Natural History 85:1-350. online.
Songa S, Liub L, Edwards SW and Wub S 2012. Resolving conflict in eutherian mammal phylogeny using phylogenomics and the multi species coalescent model. PNAS 109 (37): 14942-14947.
Spaulding M, O’Leary MA and Gatesy, J 2009.
Relationships of Cetacea (Artiodactyla) among mammals: Increased taxon sampling alters interpretations of key fossils and character evolution. PLoS ONE 4:e7062. doi:10.1371/journal.pone.0007062:1-14.

why evolution is true – mammal tree



1 thought on “Mammal phylogeny employing fossil and extant taxa

  1. Actually, there are analyses that use both molecular and morphological data and incorporate a lot of fossil taxa. Take O’Leary and Gatesy (2008) which used 43 extinct (including six fossil whales and fourteen stem hippos) and 28 living taxa, 635 morphological characters and lots of genes. They recovered the standard molecular phylogeny, with whales closest to hippos among living taxa, etc.. So it’s not simply that fossils are missing from molecular analyses.

    Also, your “traditional” mammal groupings aren’t right. Using Shoshani and McKenna (1998) as the standard example…
    – What you call afrotheres (actually uranotheres or paenungulates for that subsection) were placed as ungulates closest to odd-toed ungulates).
    – Bats were part of what you call ‘euarchontoglires’ (the addition of the Eu- was specifically to signify excluding bats), supposed to be sister to flying lemurs, then to primates.
    – The (hedgehog + mole + shrew) group was thought to be primitive- sister to everything except xenarthrans.

    More recent much better morphological analyses like Halliday et al. (2015) (680 characters and 177 genus-level taxa, of which 130 are Paleogene) find different topologies yet-
    ( (Ptilocercus plus some primates , rabbits plus rodents , tenrec plus xenarthrans) ( (elephants , elephant shrews plus hyraxes, odd-toed ungulates, even-toed ungulates plus whales) ( carnivores ( pangolins (bats , Tupaia plus flying lemurs plus some primates) )))) .
    And no, Chriacus is closest to pangolins of all recent groups there, not to bats. Funny that they also divide up tree shrews, but find the exact opposite of you, who has Tupaia close to rabbits and rodents while Ptilocercus is closest to flying lemurs and bats. Also, Plesiadapis is by primates, whales are by even-toed ungulates, and Phenacodus is nowhere near any ungulate group.

    If I were to trust a morphological analysis of mammals, that one is superior to yours in every way (number of characters (3X) and taxa (5.2X), sampling of basal members, expertise of author, etc.). Note too that when Halliday et al. constrained their trees to match the molecular tree, the Eocene Widanelfarasia and Dilambdogale (which are otherwise out by pangolins, ‘condylarths’ and such in the unconstrained tree) fall out in Afrotheria. And where are they from? Africa. That’s an example of the convergence of data I spoke of earlier- another point in favor of molecular data.

    References- Halliday et al., 2015. Resolving the relationships of Paleocene placental mammals. Biological Reviews. DOI: 10.1111/brv.12242

    O’Leary, M. A. & Gatesy, J. (2008). Impact of increased character sampling on
    the phylogeny of Cetartiodactyla (Mammalia): combined analyses including fossils.
    Cladistics 24, 397–442.

    Shoshani and McKenna, 1998. Higher taxonomic relationships among extant mammals based on morphology, with selected comparisons of results from molecular data. Mol Phylogenet Evol. 1998 Jun;9(3):572-84.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.