In the phylogenetic battle between traits and genes
author and paleontology professor Michael J Benton (2014) chose genes and lost fossils in his cladogram of placental mammals (Fig. 1). Benton writes, “The phylogeny of placental mammals has been substantially overturned by the application of molecular phylogenetic techniques.”
The resulting cladogram is a confusing mix
where university students are told that bats (Chiroptera) and horses (Perissodactyla) are close relatives (Fig. 1), among many other faults (see below). This should scare you if you plan on buying this book or paying tuition to attend Benton’s lectures.
This should also scare you away
from using suprageneric taxa in a cladogram. When you don’t use generic or species-grade taxa too much is left unsaid, leaving too large of a gap between taxa (e.g. no taxa nest between bats and whales in Benton’s cladogram, Fig. 1).
And don’t omit fossil taxa
for the same reason.
This is why
the large reptile tree (LRT) tests 1907 generic taxa (including fossils) based on traits. In the LRT bats are not related to horses, as every grade-schooler knows, but every university paleo student has to swallow and regurgitate on tests to get a good grade.
deep time analyses based on genes too often deliver false positives. Get used to it. It’s a fact of life. Gene results are based on trust with no correlates (at present) determining skull/orbit ratios, presence of a quadratojugal, etc. Benton trusts that someday a last common ancestor of horses and bats exclusive of whales and cows will be found. Good luck with that.
trait results in the LRT have no trust issues. Traits document evidence you can see and measure with calipers from snout to tail tip. Trait results deliver a gradual accumulation of evolving traits at every node of the LRT, and that includes fossil taxa.
Lets look at some other issues in Benton’s cladogram
from top to bottom:
= aardvarks. In the LRT aardvarks (Orycteropus) are naked armadillos in the clade Xenarthra.
= elephant shrews, but elephant shrews are diphyletic in the LRT. Macroscelides nests between Tupaia, the tree shrew and Chrysochloris, the golden mole. The unrelated Rhynchocyon nests with tenrecs, derived from fossil anagalids and leptictids.
= golden moles (Chrysochloris), tenrecs (Tenrec), hedgehog tenrecs (Echinops), otter shrews (Potamogale) and shrew tenrecs (Microgale). In the LRT many of these taxa nest at the bases of several trait-based clades, all with a tree shrew-like ancestor not recognized by Benton’s cladogram (Fig. 1).
Proboscidea, Hyracoidea and Sirenia
= elephants, hyraxes and manatees. In the LRT these taxa are also related, but they have a deep fossil history and are not closely related to the taxa listed above in Benton’s gene-based cladogram (Fig. 1).
= sloths, armadillos and anteaters. In the LRT these highly derived taxa are also related to each other, but they have a deep fossil history and do not give rise to elephants and golden moles. Based on skeletal traits aardvarks are xenarthrans.
Lipotyphyla = Eulipotyphla
(formerly ‘Lipotyphla’ – Afrosoricida = ‘Erinaceomorpha’ + ‘Soricomorpha’) = hedgehogs, moonrats, shrews, solenodons, and moles. In the LRT all are members of Glires (alongside shrews, rodents, rabbits, plesiadapiformes and multituberculates) in the LRT. One exception: Talpa, the mole, nests with mongooses in the Carnivora. Uropsilius, the shrew-like mole, still nests with other shrews (see below).
= pigs, camels, deer. In the LRT these taxa are also related. Hippos are not related.
= whales. In the LRT odontocetes arise from fossil pakicetids and echo-locating tenrecs. Mysticetes arise from fossil mesonychids, extant hippos and fossil desmostylians.
= bats. In the LRT bats and pangolins share a small, fossil tree shrew last common ancestor after the split from colugos. Horses and rhinos? No.
= horses and rhinos. In the LRT these taxa are related to each other, but are not related to bats.
= mongooses, bears, seals, sea lions, cats and dogs. In the LRT these taxa are related to each other (and add the mole, Talpa, see above), but are not related to horses and rhinos. This is the first clade to split from the others in the placental subset of the LRT, not one of the last (Fig. 1).
= pangolins. In the LRT pangolins are closer to bats and colugos than to carnivorans.
= tree shrews. In the LRT tree shrews are basal arboreal placentals arising from derived didelphid marsupials.
= colugos (flying lemurs). In the LRT dermopterans arise from tree shrews. So sometimes genes and traits match.
= lemurs, monkeys, apes. In the LRT these taxa are related to one another, but arise from tree shrews, not the other way around as shown in Benton (Fig. 1).
Who is to blame for this mess?
Author MJ Benton?
He borrowed gene studies from O’Leary et al. 2013 and they were published in Nature. So, if so, he was scammed. (See yesterday’s post for more of Benton’s backstory).
Professors around the world who ordered Benton 1990, 1997, 2005 and 2014?
Teaching is how professors make their living and Benton’s book is what they taught and were taught for the last thirty years. So, if so, they were scammed.
Students who bought Benton 2005 and 2014 without vocal objection?
Naive freshmen just want to learn all there is to learn about paleontology. They trust that their book and tuition monies are being spent properly. They don’t have the background to say, ‘that seems untenable.’ So, if so, they were scammed.
A wide-gamut, trait-based cladogram based on traits and including fossils,
is a powerful tool. You should have one if you’re interested in vertebrate paleontology, if you teach vertebrate paleontology and especially if you write textbooks on vertebrate paleontology.
Benton MJ 2014. Vertebrate Paleontology 4th ed. Wiley-Blackwell 480pp.