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

The Emergence of Whales, Chp. 16

Synopsis of Chapter 16: Cetacean Origins: Evolutionary Turmoil during the Invasion of the Oceans

by J.G.M. Thewissen
Department of Anatomy
Northeastern Ohio Universities College of Medicine
Rootstown, Ohio


Started with Remington Kellogg (1936): Protocetus atavus was the best (and most complete at the time) model for the “cetacean archetype”. Protocetus is “remarkably primitive”, with heterodont teeth, caudal edge of nasal opening over P^1. Still, it was “unambiguously a cetacean and separated by a large morphological gap from its four-footed terrestrial ancestors.” (EEK! A gap!) Due to Protocetus, many felt that whales had a very ancient origin and no close modern relatives. This is key: it was also though that mysticetes and odontocetes were derived independently from land mammals, as recently as Slijper in 1962.

Van Valen in 1967 postulated common ancestor for mysticetes and odontocetes, and mesonychian ancestry. Still held among morphologists, but now there is the molecular evidence of the artiodactyl link.

Fossils “only began to fill the gap between Protocetus and mesonychians in the 1970s and 1980s.” Discusses the discoveries in India and Pakistan. Now new discoveries, four earliest cetacean families (see Chapter 1). Rest of Intro puts each of the foregoing chapters in perspective.

In this chapter, he aims to provide a “heuristic outline” of the field.


2.1 Paraxonia

Definition: foot has four toes, three phalanges. Other anatomical characters given.

Paraxonia incl. Cetacea, Mesonychia, and Artiodactyla. Thewissen used Cete for paraphyletic mesonychians and Cetacea. Geisler and Luo used “Acreodi” in this book for mesonychids and Cetacea (monophyletic).

Early century — felt that it was impossible to identify cetacean terrestrial relatives. Now its held to be the ungulates. Confirmed by hooflike morphology of distal phalanx of the earliest whales. More discussion of previous work, culminating with agreement that artiodactyls are the sister group to cetaceans, with new molecular data indicating that artiodactyls are paraphyletic if cetaceans are excluded.

Work also indicates sister grouphood with hippopotamids. Milinkovitch and Thewissen (1997) indicated that if this is accurate, defining characteristics of artiodactyls needed reevaluation. In particular:

“… the unique trochlear shape of the astragalar head. If Cete (mesonychians and cetaceans) is included in artiodactyls, then the trochlea was lost in this branch, as it is absent in mesonychians, but present in the earliest artiodactyls. A complete astragalus of an early whale will shed light on this matter.”
(Remember that Thewissen has since tried to put together some fragmentary whale ankle bones, and what he was able to piece together didn’t resolve the question.) If they found a trochleated astragalus in early whales, that would clinch it. But if not: “several evolutionary scenarios are possible.” Summary of remaining here: more work needs to be done.

2.2 Cetacea

Characteristics of cetaceans given. (See Geisler and Luo for the skull stuff.)

Cetacea includes Odontoceti, Mysticeti, and the archaeocetes, which are paraphyletic (OK, I’ll type ’em one more time, but this is the last time: Pakicetidae, Ambulocetidae, Remingtoncetidae, Protocetidae, Dorundontidae, and Basilosauridae). Only the Rems and Basils have established their monophyly.

“Cetacean” used to mean obligate marine predatory mammals. Now that we have all these gap-filling taxa, “a more refined concept is necessary”. Best to include pakicetids and sister group in Cetacea (this sentence is weird). Hearing and feeding adaptations fit, but can’t yet evaluate locomotor characters. Geisler and Luo support cetacean monophyly.

2.3 Pakicetids

Thewissen and Hussain (1998) listed primitive characters of pakicetids. I am not typing all of them. But there is this: “… the degree of ossicular rotation in Pakicetus forms an excellent intermediate for the full middle ear rotation of later cetaceans (ref) and cold be considered a stage along a morphocline.”

Oldest discovered pakicetids not initially recognized as cetaceans, even in 1940s. Gingerich first to recognize them as whales, but not first to publish description (darn!) Considered most primitive family of cetaceans. It’d be nice to have postcranials.

Three genera, Pakicetus, Nalacetus, and Ichthyolestes. Pakicetus was about the size of a coyote; the other two were smaller! Teeth of Ichthyolestes indicate less abrasive diet than Pakicetus (fish?)

Gingerich apparently reconstructed Pakicetus as well adapted for locomotion in water (ca. 1983), but there’s no evidence for it. Underwater hearing possible (again Gingerich) but morphology more likely for “inertial hearing”. If so, then “it is possible that the earliest cetaceans evolved sophisticated hearing by using their mandibles as a bony interface between substrate (soil or possibly water) and middle ear.” Useful for footstep detection. More work with modern mammals necessary.

Pakicetids from shallow freshwater deposits = terrestrial / amphibious mammals. Note that mesonychians are found in the same formations.

2.4 Ambulocetids

They have two apomorphies: ptyerygoid process similar in height to the braincase, orbits face laterally close to sagittal plane. Other characters given. Sister group of all “nonpakicetid” cetaceans, and (I’d expect him to say this) “form a good functional model for the ancestral marine cetacean”.

Gives history of discoveries, early ones not recognized as cetaceans, let alone ambulocetids. Talks about his own discovery, which “made it possible to recognize that this family documents an amphibious stage in whale origins”. Not all elements of holotype have been described due to safety regions (i.e. gun-toting armies). 1996 got more vertebral column and pelvis.

Discussion of what Gingerich covered (weight estimate). Deposits and isotopes indicate euryhaline, partly dependent on freshwater.

The next paragraph expands a bit on what I commented on in Chapter 15. Tooth wear indicates it ate hard objects = large and struggling prey? Thewissen suggested crocodiles as modern analogue. “Crocodilelike forms occur as intermediates in the land-water transition of several tetrapods (ref)”. So likelihood of Ambulocetus being an ambush predator in shallow water. Localities where found have marine plants and shallow marine invertebrates.

Still would like to know: middle and inner ear morphology, and the tarsals (important for artiodactyl-cetacean relationships).

2.5 Remingtoncetids

Lists synapomorphies. 1st whales from south Asia. Postcranial material available, first attributed to protocetids. Gingerich’s work looks pretty good.

Discussion of characteristics of the members. Probably did not have descendants.

Sense organs are very different from other early cetaceans. Small orbits indicate eyes were not the primary sense organ. Better hearing is indicated, likely for underwater hearing. Remingtoncetids are the most primitive cetaceans for which the inner ear is well- known. More study will aid understanding of origination of underwater hearing.

= “Peculiar animals”. Jaw morphology indicates diet of fast-swimming, easily subdued aquatic prey, but skeleton indicates they were slow. Ambush predators specializing in small prey? Requires more firm association of limb and cranial material. Nearshore marine and littoral environments.

2.6 Protocetids

Paraphyletic group probably including the sister group to Eocene and younger cetaceans. They have a combination of primitive and derived characters. This has been covered before. Main primitive feature is a complete dental formula.

Phylogeny can’t be analyzed in “explicit” detail, because a number of specimens still have to be described in detail (got that? There’s still more of these guys!) Hulbert (the Georgiacetus man) and literature indicate three “grades”:

1. Most plesiomorphic: Indocetus, Rodhocetus (which might be Indocetus) and Takracetus.

2. Protocetus and Babiacetus

3. Pappocetus, Georgiacetus, and Eocetus

Basilosaurids and dorudontines are probably derived from protocetids. He considers them distinct at family level; others consider them subfamilies, Gingerich doesn’t. Notes that the great elongation of Basilosaurus is unique. “This difference between basilosaurids and dorudontines is much larger than the differences between many of modern cetacean clades that are considered distinct at the family level.”

Protocetid diversity highest in Indo-Pakistan; earliest cetacean family found on other continents. All had reduced hindlimbs, from what postcranial skeletal material is known. See Hurlbert’s chapter (8) for more on Georgiacetus.

Protocetids “seem firmly committed to the marine environment”. Dentition less suited to mastication (chewing). Skulls show that the endocranial vascular retia evolved in protocetids.


From the 1st paragraph: “Cetaceans originated when a Paleogene land mammal underwent a dramatic shift in biological attributes in order to accommodate an enormous shift in habitat.”


From the 2nd paragraph: “The origin of cetaceans is already better documented than most other major evolutionary transitions, and no end is in sight for the cascade of discoveries.”

Third paragraph covers the questions: physical setting (where, when, how long), biological patterns (what changed), evolutionary process (how and why). Still work to be done.

Still would like to have more postcranial anatomy for the earliest cetaceans.

Physical setting knowledge is “known to some extent”. Need more studies of terrestrial mesonychians.

Protocetids are known worldwide, and these “were probably the family that conquered the oceans”. Still need better chronology; not easy due to the complications of the Himalayan Orogeny. From pakicetid to protocetid in ~5 million years!

Discussion of evolutionary processes: basically what’s covered in the book, hearing, reproductive system, and locomotion. Form and function is “of paramount importance in the interpretations of fossil sequences”.

Why’d they get into the water? Suggested due to the “plethora” of resources in the ocean available since the extinction of Mesozoic marine reptiles. This is “simplistic”. “The earliest cetaceans lived in or near freshwater and it is unlikely that they profited from extinctions in the nearshore marine realm.” Probably competed with crocodiles. (A question he doesn’t address — maybe this competition drove them seaward?) Dentition indicates they were food specialists.

Final paragraph:

“In sum, cetacean origins are already one of the best documented examples of major morphological change in the fossil record in the fossil record (Gould, 1994). There is great promise in the study of this morphological change in a rigorous phylogenetic context with understanding of the relation between form and function, and adequate data from soft anatomy and embryology.”

BION – that’s it. I will compose a “Reflections” post on why I did this, what I learned, whether I found what I was looking for, and what I think others should take from this extended effort. In my opinion it’s been worth it. It is truly amazing what we know (and how scientist have determined it).

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

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