The Emergence of Whales, Chp. 8

Synopsis of Chapter 8: Postcranial Osteology of the North American Middle Eocene Protocetid Georgiacetus

Richard C. Hulbert, Jr.
Department of Geology and Geography, Georgia Southern University, Statesboro GA 30460-8149.
* Now with the Florida Museum of Natural History, Gainesville, FL.

I. Introduction

Let’s start with a bang. First sentence:

“Archaocetes whales are a paraphyletic assemblage of species that represent evolutionary intermediaries between modern, fully marine cetaceans (Odontoceti and Mysticeti) and their terrestrial ancestors, the Mesonychia (refs).”

First (for 80 years) was Protocetus. Then more recently, Ambulocetus, Rodhocetus, Remingtoncetus, and some others, (as we know) with more complete post-cranial skeletons than Protocetus. Hurlbert will compare the post-cranial skeleton of Georgiacetus to the others.

For ease of typing, Georgiacetus vogtlensis will be called “Gv”.

Three specimens of Gv have been found. It’s the 1st formally described North American protocetid, others are under study. It’s also one of the youngest protocetids (40-41 mya, late middle Eocene). By phylogenetic analysis, it is more closely related to the basilosaurids than older archaeocetes. It is therefore “critical in understanding the transition from protocetid- to basilosaurid-grade cetaceans.” (!!)

(For those keeping score, just in the Introduction, he has used the term “evolutionary intermediaries” and the word “transition”. No punches being pulled here.)

Next paragraph describes what was found and the unfortunate likelihood that some of it was probably destroyed and damaged during initial exposure by construction workers.

2. Materials and Methods

Descriptions are based on direct observations of the Gv specimen at Georgia Southern, and Protocetus in a museum in Stuttgart.

Vertebral sequencing based on best fit; obviously some are missing.

3. Postcrania of Gv

3.1 Vertebrae 3.1.1 Cervial vertebrae

Three found. C4 and C7 in good shape, C5 was damaged. Skip the morphology (but for Scrabble players, don’t forget “hypapophysis” if you have a lot of ‘P’s, also the pre- and postzygapophysis).

3.1.2. Thoracic vertebrae

Nine of the most likely number of thoracic vertebrae (13) have been found, based on two of the specimens.

[I should note that the article contains several photographs of the vertebrae.]

He describes the mis-fits that indicate missing vertebrae. Morphology makes it fairly easy to distinguish the anterior thoracics from the posterior thoracics.

3.1.3 Lumbar vertebrae

Five of the likely number (8) have been found. More morphology of each. The detail here is astonishing: I’d bet that Hurlbert knows each of these bones better than the lines on the palm of his hand.

3.1.4 Sacral vertebrae

Four are described, S1 to S4. He doesn’t say if any are believed to be missing. Only one of the four is significantly damaged.

3.1.5 Caudal vertebrae (that would be the tail)

Only one (which articulates posteriorly with S4).

3.1.6 Hemal arch

There is one, also called the “chevron bone”.

3.1.7 Ribs

The best specimen of the three (GSM 350) included seven left and five right ribs.

3.4 PELVIS

Both innominates (left and right) are found from GSM 350. Left side is more complete, right side has details of the acetabulum (Scrabble alert). The acetabulum happens to be large, with a deep circular articular surface that opens widely postventrally. [This is the type of thing that I’ve been skipping.] Note that muscle attachments can be discerned as well.

OK, for a long chapter we’ve covered a lot of pages thus far. But now it gets more interesting.

4. Comparisons between Gv and other archaeocetes

4.1 Introduction

Let me just type in and savor the opening paragraph. I think there’s a message here:

“During the Eocene, cetaceans evolved (!!) from semiaquatic, fluvial, relatively small-bodied mammals capable of both terrestrial and aquatic locomotion (!!) into much larger, highly specialized, fully marine animals incapable of walking on land. Adaptations (!!) to the postcranial skeleton were an essential component of this transformation (!!). The fossil record of archaeocetes, although still incomplete, demonstrates the stepwise nature (!!) and rate (!!) of this transformation. _Georgiacetus_ represents a grade of cetaceans far along in this evolutionary series (!!!!!!!!), one in which adults probably spent nearly all of their lives in the ocean, returning briefly to land to give birth. While on land they were incapable of standing erect or walking on their limbs, but instead must have relied on undulations of the body.”

In this section, the postcranial skeleton of Gv will be compared to what’s known of the others, which are Ambulocetus, Remingtoncetus, Dalanistes (a remingtoncetid), Rodhocetus, Gaviacetus, and Protocetus (protocetids), Zygorhiza, Dorudon, and Ancalacetus (dorudontines), and of course, Basilosaurus. Gv looks most like the protocetids, because it is one.

4.2 Vertebrae

Mesonychids: 12 thoracic, 7 lumbar, 3 sacral vertebrae

Remingtoncetids: four fused sacral vertebrae

Rodhocetus: 13 thoracics and 6 lumbar vertebrae

Habib Rahi whale (found in 1991 by Gingerich, not yet described, older than Rodhocetus): also 13 thoracics, last is diaphragmatic

Gv (conservative best estimate): 13 thoracic, 8 lumbar, 4 sacral, with two postdiaphragmatic thoracics. ,p> SO… later archaeocetes increased the number of lumbar vertebrae and the number of postdiaphragmatic presacral vertebrae, AND increased the functional length of the lumbar region by increasing the lengths of the individual centra and by freeing the sacral vertebrae (!!), which allowed the sacral vertebrae and the anterior caudals to function with the lumbars as a unit. Gv has longer lumbar and sacral vertebrae than predecessors.

Cervicals of Gv are most similar to Rodhocetus and Protocetus (ambulocetids and remingtoncetids have long centra in the cervicals, protocetids relatively short).

Thoracics of Gv are similar to Protocetus, but larger and more robust

The sacrals (this is IMPORTANT)

First of all, the likelihood of a fused sacrum is inferred for Ambulocetus due to the robustness of the hind-limb elements. Remingtoncetids had a fused sacrum, so a similarity is likely. Then:

“Interestingly, the four known examples of protocetid sacrals are all slightly different, forming an exemplary morphocline (!!) between the fused, plesiomorphic state of remingtoncetids and the condition observed in basilosaurids.” … (detail) …

“The morphology of the sacral vertebrae in protocetids forms a continuum (!!) that spans the differences observed in remingtoncetids and basilosaurids.” Detail follows. I will say that if a schematic diagram of the changes could have been made, it would have been a nice contribution here, but the paper is written for professional paleontologists, not a lay audience, and they likely know what all the features actually are.

Not commonly found. Ambulocetus had a few, Rodhocetus had a relatively complete rib cage but it hasn’t been described yet. (Gingerich is very busy.) Comparing Ambulocetus to Gv, Ambulocetus had more robust ribs with greater curvature. The distal ends of Gv’s ribs are morphologically similar to basilosaurines and Zygorhiza, but the features in the former (Gv’s ribs) are not seen as prominently.

4.4 Pelvis

None for pakicetids or ambulocetids. Some partials for remingtoncetids, featuring “broad articulation” between the ilium and sacrum. So the other good pelves are for Rodhocetus and Basilosaurus. Overall morphology and size of the protocetid pelves is that of typical terrestrial mammals, compared to basilosaurid pelves. Although the sacral vertebrae of Gv are like basilosaurids (loss of iliosacral articulation), the pelvis of Gv “is surprisingly plesiomorphic in appearance”. (To review, plesiomorphic means a primitive character state that is derived from the ancestor without change. I.e., the pelvis of Gv still looks a lot like that of a terrestrial mammal.)

Comparing Gv to Rodhocetus, Gv has
– a relatively longer ilium
– a shorter, more reduced ischium
– a well-developed acetabular notch
– a much wider pubis

“Overall, the younger Gv possesses more derived character states throughout its skeleton related to aquatic adaptations (!!) relative to Rodhocetus.” But the longer ilium and acetabular notch are plesiomorphic, suggesting that there was more than one protocetid lineage in the Eocene (geographic separation?) each independently adapting to a marine way of life. (!!) The dissimilarity of the two pelves of Rodhocetus and Gv imply different styles of aquatic locomotion.

5. Conclusions

Gv is the youngest, most aquatically adapted protocetid that has postcranial elements and a skull. Increase in the postdiaphragmatic vertebral section (an archaeocete trend) is demonstrated in Gv compared to Rodhocetus four ways. The changes in the sacral and anterior caudal vertebrae allowed them to function more like lumbars, and a necessary consequence was “loss of direct articulation between the pelvis and sacrum.” This made movement on land more limited, and would also have caused a decreased use of the hindlimbs in aquatic locomotion. As previously described, cetacean aquatic locomotion evolved through four major stages in the Eocene: (1) quadrupedal paddling (2) dorsoventral undulations using hind feet as hydrofoil, (3) shift of hydrofoil to fluked tail, and (4) total dependence on tail, complete loss of locomotor function in the hind limbs.

“Gv neatly (for the most part) fills the morphologic gap (!!) in this sequence between early protocetids (such as Rodhocetus) and basilosaurids.” I told you I liked his style.

The plesiomorphy of the Gv pelvis indicates that the hind limb may have retained some limited locomotor function.

Fantastic.

After all that, Chapter 9 is bound to be somewhat of a letdown, but we shall continue:

“Homology and Transformation of Cetacean Ectotympanic Structures” by diagram champion Zhexi Luo