table of contents
September 2001
[back] [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [next]

The Emergence of Whales, Chp. 3

Synopsis of Chapter 3: Molecular Evidence for the Phylogenetic Affinities of Cetacea

John Gatesy, Laboratory of Molecular Systematics and Evolution Department of Ecology and Evolutionary Biology University of Arizona, Tucson, AZ

Let me state this right upfront: I don’t understand most of this molecular genetics material. Again, if I find it of particular interest, I’ll mark it with (!!).

And this chapter provides much fodder for anti- evolutionists, especially if they pull quotes out of context. Like this one from the 1st paragraph:

“The rapidly growing molecular data base should, in theory, complement anatomical evidence from the spectacular fossil discoveries of the past 15 years. Unfortunately, recent phylogenetic analyses show more conflict than compromise between molecules and morphology.”
But before the anti-ev forces exult too much, what does he say next?
“There is a marked discontinuity in morphology between extant cetaceans and their closest living relatives, but recently described fossils have closed this anatomical gap. Gross anatomical data clearly demonstrate that whales are highly transformed ungulates [refs, incl. subsequent papers in the book]. “Mesonychians” from the Paleocene and Eocene are considered the closest terrestrial relatives of Cetacea [refs] and various “archaeocete” genera compose the stem lineage of modern whales [more refs].”
Intro to the problem: Cladistic analyses put Cetacea closer to Perissodacytla and Paenungulata (Sirenia + Hyracoidea + Proboscidea) than to Artiodactyla. BUT, other interpretations suggest sister group relationship between Artiodactyla and Cetacea, more consistent with molecular results.


“Oddly, many molecular results insert Cetacea within Artiodactyla [refs] and some genes specifically link cetaceans with hippopotamid artiodactyls [refs]. These results contradict morphological evidence in support of artiodactyl monophyly [refs] and demand extensive gaps in the fossil record of the Paleocene or remarkable evolutionary convergence between mesonychians and early cetaceans [ref]. Sharp conflicts between molecules and morphology, as well as incongruence among different molecular trees [ref] have led to skepticism of the molecular evidence for phylogeny.”
So what Gatesy is going to attempt is to summarize previous results and then use new DNA sequences to (hopefully) place Cetacea relative to the other extant ungulate orders, assess artiodactyl paraphyly, and test the putative sister-group relationship between Hippopotamidae and Cetacea. Later on he says he’s going to do the following, which is similar but also frustratingly different:
  1. Relate Cetacea to the five other extant ungulate orders
  2. Assess the monophyly of Ungulata
  3. Test whether Cetacea is a sublineage of Artiodactyla
  4. Determine which of the five major extant artiodactyl clades is most closely related to Cetacea (if Cetacea is indeed a sublineage of Artiodactyla)
Summary of previous results:

There have been five previous molecular data used to examine cetacean phylogeny:

  1. Immunological reactions
  2. DNA-DNA hybridization
  3. Amino acid sequences
  4. DNA sequences
  5. Retroposons
His next six figures show sixty trees from 44 previous molecular studies, spanning 50 years (!!).

General pattern summary:

  1. Artiodactyls are the closest relatives of Cetacea in the majority of trees. No other eutherian order is consistently resolved as the extant sister group to Cetacea.
  2. Perissodactyla often groups with Artiodactyla and Cetacea to the exclusion of other ungulates.
  3. A monophyletic Ungulata is inconsistent with most molecular evidence.
  4. Artiodactyla is often paraphyletic with Cetacea clustered as an artiodactyl subclade.
Further work supported a Hippopotomidae – Cetacea sister group relationship. (!!)

So how is he going to do this:

With seven genes:
gamma-fibrinogen (blood coagulation)
Protamine P1 (nuclear condensation during spermatogenesis)
12 S and 16 S rRNAs (mt ribosome)
kappa- and beta-casein (milk protein genes)
Cytochrome b (oxidative phosphorylation system)

From HOW many different species?
As many as 67. 19 taxa were sequenced.

Sequences aligned with MALIGN, “a multiple sequence alignment program that uses parsimony as the basis for alignment choice”.

Phlogenetic analyses used PAUP 3.1.1. and PAUP*4.0d55

— There is a quite a bit of detail on methodology that is completely lost on me. So we’ll have to content ourselves with the results.


First he talks about differences between the different kinds of analyses. Stuff like this:

“Protamine P1 and mt 12S + 16 S rDNA are incongruent with four and five of the other datasets, respectively. Because the ILD test results are not transitive, decisive boundaries between “congruent” and “incongruent” character sets do not exist.”
I’m sorry I can’t be of much help with that. So let’s move on to the results I can understand:

A. A Cetacea-Artiodactyl clade is supported “clearly”, “with high to moderate support”, or “weakly suggested”, depending on the gene. Perissodactyla is the next closest ungulate relative.

B. There is little support for a monophyletic Artiodactyla. “No groupings of artiodactyls are well supported by the combined mtDNA data set (Fig. 16) but Cetacea + Hippopotomidae is favored when the mtDNA are weighted by character fit [ref].”

C. The weight of DNA evidence favors a Hippopotomidae + Cetacea clade, a Hippopotomidae + Cetacea + Ruminantia clade, and a Hippopotomidae + Cetacea + Ruminantia + Suina clade with Camelidae more distantly related.

D. The combined molecular evidence places Cetacea as a subclade of a paraphyletic Artiodactyla. (!!)

Gatesy then discusses the morphological evidence supporting this. I’d love to type in all the synapomorphies for Artiodactyla, but my fingers are tired. Still:

“Among these [the list I didn’t type], narrow trigonids, enlarged lacrimals, absence of alisphenoid canal, and a paraxonic foot are characteristic of the earliest cetaceans [refs].”
The next paragraph is quite interesting, but if someone could show me a picture of a trochleated distal astragalus, I would really be grateful.
“The classic artiodactyl ordinal character, a trochleated distal astragalus [ref] has not been observed in cetaceans. However, the search for a cetacean astragalus has been frustrating. The bone is fused to the other tarsals in the vestigial hind limb of Basilosaurus [ref], the distal end of the astragalus is missing in Ambulocetus [refs], the astragalus is unknown from all other extinct whales, and it has been lost along with most of the hind limb in extant whales. Because the entire tarsus has been lost within Cetacea, it is illogical to assume that a trochleated distal astragalus could not have been lost within “Artiodactyla”. ”
So, did you get all that? Well, maybe, but here’s one more stab at it:
“Past molecular hypotheses of cetacean phylogeny are diverse, but four common features of these studies stand out:
A. Cetacea often groups with Artiodactyla
B. Perissodactyla is closely related to Artiodactyla + Cetacea.
C. Ungulata is often diphyletic.
D. Artiodactyla is often paraphyletic.

2. These four common features of previous molecular studies are supported by phylogenetic analyses of DNA sequences from four nu and three mt genes. A cladogram based on over 4500 aligned nucleotide positions suggests Cetacea is nested three nodes within a paraphyletic Artiodactyla. In sum, the DNA evidence supports a sister-group relationship between Cetacea and Hippopotomidae.

3. Trees derived from subsets of the total DNA data base are contradictory, and there are several examples of significant character incongruence among data partitions. However, specific topological discrepancies are not well supported in most instances.”

After all that, I’m pretty clear on one thing at least: according to the molecular data, the closest existing relative on land to all the whales is the hippopotamus.

[back] [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [next]

This page was last updated September 1, 2001.
It was reformatted and moved August 6, 2007
Copyright © 2001 by James Acker

table of contents
September 2001