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. 14

Synopsis of Chapter 14: Isotopic Approaches to Understanding the Terrestrial-to-Marine Transition of the Earliest Cetaceans

Lois J. Roe, J.G.M. Thewissen, Jay Quade, James R. O’Neil, Sunil Bajpal, Ashok Sahni, and S. Taseer Hussain

LJR: Div. of Ecosystem Sciences, UC-Berkeley
JGMT: Dept. of Anatomy, NE Ohio Universities College of Medicine, Rootstown, Ohio
JQ: Desert Lab. and Dept. of Geosciences, Univ. Arizona, Tucson
JRO: Dept. of Geological Sciences, Univ. Michigan, Ann Arbor
SB: Dept. of Earth Sciences, Univ. Roorkee, Uttar Pradesh, INDIA
AS: Center for Adv. Studies in Geology, Panjab University, Chandigarh, INDIA
STH: Dept. of Anatomy, Howard Univ. College of Medicine, Washington DC


This is a good and intriguing first two sentences:

“The fossil record is replete (!!) with examples of evolutionary transitions (!!) between marine and freshwater environments, in both directions. Perhaps the most striking and best documented (!!) example of such a transition is the evolution of cetaceans (whales, dolphins, and porpoises) from the extinct group of terrestrial mammals called mesonychians.”
Then they say where: the Tethyan epicontinental sea in the zone of convergence between the Indian plate and southern Eurasia.

When: during the early stages of the continental collision that eventually produced the Himalayas.

(All in the first paragraph.)

Deriving cetaceans from land mammals needed profound changes in anatomy and biology. Much previously covered. BUT: morphology is only part of the story. Becoming fully marine and _independent_ of freshwater requires physiological evolution, particularly the excess salt load associated with seawater ingestion, either when drinking water or when eating.

Most cetaceans never enter fresh water: Phocoena phocoena can enter brackish water with low salinity, and there are four (rare) freshwater dolphin species (which evolved from marine species).

Q: When did the cetacean osmoregulatory system become able to handle the excess salt load? Key to dispersal in mid-Eocene.

How: Stable C and O isotopes. Use to answer:

  1. When did cetaceans become fully marine?
  2. Was the transition to marine life gradual or abrupt?
  3. Was the transition to seawater ingestion synchronous with transition to a marine diet, or did one precede the other?
  4. How much did early whales differentiate ecologically?
Using teeth and bones from modern and Eocene cetaceans. They will also evaluate the effects of diagenesis.


2.1 Ion and Water Balance in Modern Cetaceans

2.1.1. Overview

(May I note right here that this chapter has a ton of text.)

All mammals lose water via a variety of processes. Kidneys in most mammals cannot concentrate excess salt strongly, causing death by dehydration.

Still unclear how cetaceans do it. Some identified mechanisms:

  1. osmotically driven water transfer across integument (skin/hide);
  2. active drinking
  3. metabolism of food/body fat to produce freshwater
  4. concentration of ions in urine to conserve water
Also, concentration of lipids in milk (??), lower respiratory rates, undersaturation of water in expired air.

Kohn (1996) indicates O isotopes will be dominated by transcutaneous flux. Thus, the O isotopes would reflect the environment in which cetaceans live (not drinking water, diet or species effects.). Broadly applicable?

Once thought cetacean skin impermeable to solutes and water. Some recent studies indicate water can enter, excluding sodium. Contrast with previous studies on drinking, urine excretion, and metabolism. There’s still some questions here.

2.1.2. Understanding Cetacean d^18O(sub p)

Note: I can’t reproduce the isotopic symbology in ASCII text. d will stand in for Greek delta. %o stands for parts per mil.

Since they aren’t sure whether the environment or drinking/diet is the main influence, they will primarily determine when cetaceans became independent of freshwater and able to eat marine food.

2.2 Oxygen Isotope Variations in Fresh and Marine Waters (including a quick review of stable oxygen isotope geochemistry)

The quantity being examined is the ratio of O-18 to O-16. This ratio varies in fresh and marine water due to kinetic and equilibrium fractionation in the hydrologic cycle.

Simple explanation on this page: http://palaeo.gly.bris.ac.uk/communication/Willson/isotopeevidence.html

Very detailed explanation on this page: http://wwwrcamnl.wr.usgs.gov/isoig/isopubs/itchch2.html

Summary: Atmospheric water vapor is significantly depleted in O-18, and rainwater and river water are therefore somewhat depleted. The quantity d^18O represents the deviation, in parts per mil of the O-18/O-16 ratio of a sample compared to the ratio in standard mean ocean water (SMOW), which has d^18O = 0. So if the ratio is less than in SMOW, d^18O is negative. There are other effects, including geography, which doesn’t affect here because all the samples are from the same region.

Elevation increases the freshwater/saltwater difference, so no knowledge of paleoelevation is necessary.

Atmospheric water vapor d^18O is -13%o, and rain/river water is about -3%o. Small islands show rain d^18O ranging from -1.8 to -3.5%o. So there’s a significant and persistent offset.

To examine this in the fossils, they’ll look at bones and teeth; specifically the oxygen incorporated in the phosphate (PO4^2-). Will it work? Potentially; there’s a 3-7 %o difference between marine and freshwater cetaceans. So they want to find a 2-3 %o difference to indicate marine vs. freshwater.

2.3 Carbon Isotope Ratios (C-13/C-12) of Fresh and Marine Waters

Stable C isotopes in freshwaters are also depleted in C-13 compared to ocean waters. So carbonates can also indicate marine and freshwater environments, and this includes teeth and bones. Influence of diet has been demonstrated for terrestrial animals, and in bone collagen (but not teeth and bone carbonate) for aquatic vertebrates. In modern ecosystems, C isotope differences for marine and freshwater animals evident in soft tissues, but seagrass/terrestrial Csub4 plants overlap.

Analyzed modern cetaceans tooth and bone C isotopes, compared to fossils. Also looked for pedogenic carbonates in Miocene herbivores to look for Csub4 plant influence.


3.1 Samples

3.1.1 Modern Cetaceans

18 specimens, from small (Tucuxi dolphin) to beeeg! (Physeter, the sperm whale). Also covered different latitudinal zones, i.e. temperate to hot. Didn’t get a beluga or narwhal, apparently. Used common species when possible due to necessary destruction of sample for analysis. Also sampled freshwater dolphins. (Here’s a tidbit: Lipotes vexilifer is a freshwater dolphin from Tung Ting Lake in the Hunan Province of China. I DID NOT KNOW THAT there were any in China. Way cool. It’s funny because I was looking at a map of China very recently and I noticed the big inland lakes there for the first time. As one might expect, this dolphin, the Yangtze river dolphin, is extremely endangered. Based on what I read below, it may be the most endangered mammal on Earth right now.)

3.1.2. Eocene Cetaceans

No surprise; they analyzed fossil cetacean samples. Included Icthyolestes, Pakicetus and Nalacetus (I thought Nalacetus was only known from one tooth — don’t tell me they destroyed one tooth! I doubt it.) Also looked at Ambulocetus, Attockicetus, Andrewsiphius, Gaviacetus, Indocetus, and Remingtoncetus, and our friend Georgiacetus vogtlensis.

3.2 Analytical Methods

Obtained samples from teeth by dental drill; enamel and dentine separately if possible. Sample treatment description follows. Note that 1M acetic acid was used to remove secondary nonstructural carbonate that could confuse the analysis due to diagenetic over- printing. (See, these guys do know what they’re doing, Dr. Austin.) Tooth enamel is best to analyze because it’s most resistant to diagenetic effects. They looked at some herbivores (anthracobunids and brontotheres) to test for diagenetic effects.

C and O extractions used phosphoric acid, producing CO2 analyzed with a Finnigan MAT Delta-S series 251 gas source mass spectrometer at BOTH the U of Michigan and U of Arizona Stable Isotope Laboratories. Uncertainties were +/- 0.06% for large samples, +/- 0.1% for small. Results reported in delta (%o) notation, compared to SMOW for oxygen and PDB (PeeDee Belemnite, of course!) for C. Results will be presented as a range of values for each taxon.


4.1 Oxygen Isotope Compositions of Tooth and Bone Phosphate

4.1.1 Modern Cetaceans

Range is +10.8 to +19.6, with a 2.4%o gap between the freshwater and marine species. Freshwater: +10.8 to +15.7. Marine: +18.1 to +19.6. Similar to a previous study.

4.1.2. Eocene Cetaceans

Almost identical range. Nalacetus, Pakicetus: +15.0, +16.6 Icthyolestes: +13.8 to +16.3 %o. 4 out of 10 values are in the modern freshwater range, while 6 of 10 values are higher. Ambulocetus: big range, 13 values from +13.0 to +21.0. Gandakasia (ambulocetid from an earlier study) was +14.0 to +17.9. Attockicetus: +16.6 (one value).

Gaviacetus, Indocetus, and Remingtoncetus (Harudi formation): +18.2 to +21.8. 1.5 to 5.2%o higher than the earlier archaeocetes (lower Kuldana formation).

4.2 Carbon Isotope Compositions of Tooth and Bone Carbonate

4.2.1. Modern Cetaceans

Analyses demonstrated a difference between marine and freshwater cetaceans. Freshwater: -17.1 to -11.6. Marine (with 1 exception): -10.9 to -7.8. One specimen of Physeter (sperm whale) was -13.0, the other was -8.0. Due to migration/ingestion of highly negative (del^13C) phytoplankton? Phytoplankton carbon varies by 5%o, most negative at high latitudes.

4.2.2 Eocene Cetaceans

Earliest cetaceans (Kuldana) are different than “slightly” younger cetaceans from the Kohat, Harudi, and McBean formations. (Georgiacetus is the “McBean” formation — I wish they’d get on the same page, nomenclature-wise!) Earliest (Kuldana) whales are -14 to -12, which is in the modern freshwater cetacean range. Mesonychian was -10.1, similar to other Kuldana herbivores. Upper Kuldanans (Gandakasia and Ambulocetus) were about -14 to -10.6, 7 of 9 specimens in the -14.2 to -12 range, i.e. same as the lower Kuldana.

Attockicetus (Kohat), Gaviacetus, Indocetus, Remingtoncetus (Harudi) and Georgiacetus (McBean): -10.7 to -5.2, just like the modern mariners. Andrewsiphius the exception at -13.4.

4.3 Diagenetic Assessment of Bone, Dentine, and Enamel d^13C

Compared d^13C in bone, dentine, and enamel. Showed differences, not all in the same direction. Dentine is most significantly different, about 1%o higher. Bone closer to enamel. Largest differences found for Ambulocetus and Gandakasia in the upper Kuldana.


Authors expected temperature/isotopic composition differences between Eocene and now to affect results. Thus, the match between the Eocene and modern marine vs. freshwater specimens d^18O was notable. Georgiacetus is the main outlier, tempting to explain due to ocean cooling (as record in foraminifera). Teeth aren’t expected to record temperature changes due to formation essentially in vivo. But this trend has been observed before. Teeth are exposed to the external environment more???

Same for d^13C, a reasonable match between modern and Eocene ranges. The bone carbonate values also can be predicted from collagen values by adding 7%o, and the prediction from previous collagen studies is borne out by these bone carbonate results (modern).

Now for the shifts. From the oldest to the youngest specimens (based on the formations where they were found), there is a positive trend, i.e., the negative values become less negative (d^13C) and the positive values become more positive (d^18O). They make bivariate plots for the data. Four regions are defined:

  1. Eats terrestrial/freshwater food, ingests freshwater (primarily freshwater existence)
  2. Eats terrestrial/freshwater food, ingests seawater
  3. Eats marine food, ingests freshwater
  4. Eats marine food, ingests seawater (fully marine)
Works because ranges are mostly nonoverlapping (that Physeter sample is the only problem).

With that one exception, all of the modern cetaceans are either category 1 or 4. But for the Eocene cetaceans, there are specimens in each of the four categories. This distribution “suggests the transition from terrestrial to marine was not a simple progression, but rather mosaic in character and that some of the early cetacean species had ecological and physiological requirements that could not be inferred from morphology and depositional environments.”

Gingerich apparently suggested that the “remnant Tethys” was rich in nutrients and therefore food for Pakicetus. But the isotopes indicate that the 3 lower Kuldanans (Ichtholestes, Nalacetus, Pakicetus) had a terrestrial/freshwater diet and ingested freshwater. Alternatively, they were ingesting low d^13C food, and the change reflects an isotopic shift. Deemed unlikely because the lowest possible d^13C from marine plankton diet is still 2-3%o higher than the lower Kuldanan values, and d^18O indicates freshwater restriction.

Mesonychians: on the marine/freshwater border. Need more fossils!

Upper Kuldanans (Ambulocetus, Gandakasia): terrestrial diet, ingestion of waters with a “wide variety” of isotopic compositions. = euryhaline physiology, consistent with the nearshore depositional environment. These fossils appeared to be most altered postmortem, but diagenesis is discounted to explain low d^18O. Explanation and ideas for further testing follow.

Attockicetus: marine food, freshwater ingestion. Much different than its “close” relative, Remingtoncetus.

Gaviacetus, Indocetus, and Remingtoncetus (middle Eocene): fully marine existence is indicated. (At this point I’m going to point out that Indocetus is considered very similar to the point of being the same species as Rodhocetus, and Rodhocetus was found in deep water deposits. More and more data indicating that Rodhocetus is a superb example of a transitional fossil.)

Next paragraph is more about diagenetic effects.


Summarizes the above. Notes speculation that the mesonychian from the lower Kuldana could have had a diet combining terrestrial and marine prey. (This makes me wonder about where Kodiak bears would test out: they eat a lot of terrestrial food, but they also eat a lot of salmon.)

Finally: “The evolutionary transition of cetaceans from terrestrial to marine life was thus geologically rapid. The transition to life in seawater involved changes both in osmoregulatory physiology and diet, but the changes were not strictly coupled. The apparent decoupling of food and water requirements may have facilitated niche differentiation, and as a result, the diversification of the earliest cetaceans.”

[I find that very interesting. It tends to confirm that the move into the oceans was at first behavioral, and not (at first) physiological. That makes sense, that the adaptations would be driven by the environment, not that the organisms would first exhibit marine adaptations and then get into the water because it was better suited to them! This also indicates that animals can compensate behaviorally to an extent for physiological limitations. I.e., think of Rodhocetus as an oversized deep-diving otter.]

[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