mine what the time was when they switched from freshwater to seawa-
ter?”
“Sure. There are a number of conditions. You need an associated
fauna so you can study the context, you need to know body size, you
need modern analogues, you need to—”
Lois is about to go off on a complicated disclaimer, but I do not want
to lose the momentum, so I interrupt. “I think that we can get all those
things. How big a sample do you need?”
“Well, you need about five grams, and it would be good to have
tooth enamel, dentin, and—”
120 | Chapter 9
“That does not mean anything to me. How big a piece of bone is
that?”
“It depends on the thickness.”
“How heavy is a fingernail clipping?”
“I don’t know. I will need more than that.”
My turn to get slightly miffed. I would like to move beyond distract-
ing details. I want to know how many teeth need to be sacrificed for
this, but Lois will not be drawn into vague analogies. It would be amaz-
ing if we could track the shift from freshwater drinking to seawater. To
determine such an important evolutionary change—who could have
thought that you could figure that out from fossils? Whales obviously
would be unable to travel great distances across oceans if they needed a
freshwater source, so the ability to drink seawater may have been a
seminal moment allowing them to disperse across wide oceans.
Eventually, we decide that this is worth a try. I will take some enamel
samples, and send them to her. She will chemically pry the oxygen atoms
out of the enamel of the tooth, and lock them into a much larger mole-
cule in the same ratio of 16O and 18O as was present in my teeth. Then
she will fire those molecules through one of the arms of the mass spec-
trometer, and determine the proportion of abundance of the two mole-
cules. That will allow us to see whether that ratio is closer to freshwater
or seawater. To see whether the theory that we know of actually matches
the real world, she will also run some enamel from modern oceanic
whales and dolphins and compare that to the few species of dolphins
that spend all of their time in rivers.
drinking and peeing
Lois sends me reams of background data on the method. Although
excited, I am also worried. Is this really going to work? I am still skepti-
cal in spite of my conversations with Lois. Can such fleeting behavior as
drinking really be gleaned from these fossils?
In the meantime, I study what is known about modern marine ani-
mals and the drinking of seawater. For a thirsty animal, the problem
with drinking seawater is that there is a lot of salt in it. In fact, it con-
tains more salt than the blood and body fluids of a mammal do. As a
result, if an animal drinks seawater to hydrate itself, it needs to take
some of the salt out and excrete it, so that the saltiness of the new water
matches that of its body fluids. In birds and crocodiles, there is a gland
near the eye where salt is excreted. Mammals never have such a gland.
The Ocean Is a Desert | 121
Land mammals lose salt when they sweat, but when they live in water,
they can’t sweat, since the process is driven by evaporation from the
skin. The organs responsible for salt excretion in a cetacean, then, are
the kidneys.
To take salt out of ingested water, the animal has to dissolve it in the
urine it excretes by making that urine saltier. So the concentration abil-
ity of the kidneys is crucially important. Many small mammals, such as
mice, can excrete highly concentrated urine and therefore can drink sea-
water.1 Human kidneys cannot concentrate urine that strongly. In fact,
for human kidneys to remove enough salt from seawater to match
human body fluid, a lot of water is needed. That amount of needed
water is greater than the seawater from which the salt is extracted. A
human who drinks a batch of seawater will lose more water peeing out
the salt in that batch than was gained drinking that batch. For human
kidneys, the ocean is like a desert: there is no potable water.
Marine mammals cope with an absence of freshwater in different
ways. In a rather nasty experiment, a sea lion named Dave was locked
in a cage and only given seawater to drink, and his food was laced with
salt pills.2 Dave knew better than to drink the water, realizing that he
would lose more of his precious body water excreting the salt. For more
than a month, the animal did not drink at all, and seemed in reasonable
physical health, until, mercifully, the experiment was stopped. Appar-
ently, sea lions can withstand such dehydration. In contrast, if you hang
a running hose into the ocean in Florida, manatees may swim up to
drink from it. In spite of living in the ocean, they need a source of fresh-
water. At the other extreme, sea otters along the Pacific coast can drink
seawater freely.3 Cetaceans cannot concentrate urine to the levels where
seawater drinking becomes an option.4 Although they are known to
ingest some seawater,5 they get most of their water from their food, and
use water very sparingly.
fossilized drinking behavior
So, the isotope method may be able to answer the question of when
cetaceans learned to cope with the absence of freshwater. Lois first runs
the modern samples: pieces of teeth of some marine dolphins, a killer
whale, and a sperm whale, as well river dolphin samples from the Ama-
zon, Ganges, and Yangtze Rivers. To my delight, there is a consistent
difference between marine and freshwater species,6 and it matches what
we predicted (figure 36).
122 | Chapter 9
MODERN CETACEANS
18
habitat
oxygen isotope values ( O p)
oceanic dolphins ( n = 11)
killer whale ( n = 1)
marine
sperm whale ( n = 2)
Indian river dolphin ( n = 1)
freshwater
Chinese river dolphin ( n = 1)
South American river dolphin ( n = 1)
Freshwater
Marine
EOCENE CETACEANS
sediment
signatures
signatures
protocetid whale ( n = 3)
ocean floor
remingtonocetid whale ( n = 2)
Attockicetus ( n = 1)
Ambulocetus ( n = 8)
coastal
pakiceti
d whales ( n = 11)
riverbed
figure 36. Oxygen isotope values for modern and Eocene cetaceans.
Known habitats for the modern species indicate that isotope values can
be used to determine whether they are freshwater or marine animals.
Oceanic dolphins, killer whales, and sperm whales live in seawater, and
their oxygen isotope values are high. River dolphins from different
continents all have lower values. Thus, isotope values (indicated by
isotope geochemists by δ, in which the ratio of the two isotopes is
compared to a reference sample) can be used to identify water
ingestion behavior for the fossil Eocene species. The isotope values for
the fossils are consistent with evidence from the sedimentology of the
rocks they are found in (ocean floor, coastal, or riverbed). This plot is
based on data from Roe et al. (1998), and these results were confirmed
and refined by more modern data from Clementz et al. (2006).
Protocetid whales will be discussed in chapter 12.
Now the fossil work starts. My heart cringes every time that I have
to break a piece off a fine fossil tooth to get an isotope sample. I worked
so hard to get those teeth and not damage them, and now I am taking a
screwdriver to their shiny enamel. I put the pieces in little vials and mail
them off to Lois, who will grind them to powder. Lois sends me data
and patiently explains what all the numbers mean.
“For the pakicetids, there is a clear freshwater signature,” she says.
“Signature?”
“Signature means that the implication of the delta 18O value is that it
was freshwater.”
I know about the delta bit—it is basically the ratio of the 18O and 16O.
A lower delta value indicates more of the lighter isotope. Cool, but not
surprising: they lived in freshwater, and that is what they were drinking,
like any self-respecting land mammal. The Indian whales, such as the
The Ocean Is a Desert | 123
Odontoceti
Coast
Seas and
(toothed whales,
Mysticeti
oceans
includes dolphins
(baleen whales)
Hippopotamus
and porpoises)
modern
Rivers
and
lakes
40 million
basilosaurines
years ago
dorudontines
basilosaurids
Indohyus
ambulocetids
protocetids
50 million
pakicetids
remingtonocetids
years ago
Other
Water ingestion
Artiodactyla
(even-toed
Freshwater
ungulates)
Seawater
figure 37. Branching diagram showing the relationship between fossil whales and
artiodactyls, and the habitats in which they lived (shades of blue for water, white for
land). Water ingestion behavior is indicated as the shades of gray of the boxes with
names, and is based on isotope data. Protocetids will be discussed in chapter 14, and
Indohyus is an even-toed ungulate that will also be discussed in chapter 14.
remingtonocetids, are on the marine end of the scale. Not just cool, but
also surprising. Those whales lived in the sea near the coast, and their
isotope values show that they are independent of freshwater altogether.
This means that within a few million years of entering the ocean, whales
did not require a freshwater source and could travel across oceans
(figure 37). That is an interesting contrast with modern manatees. They
originated around the same time as whales, but they still haven’t figured
out how to live on seawater only.
Of course, history matters here. Whales and manatees are derived
from different land ancestors, and isotopes only show you what an ani-
mal drank, not what it would be able to drink if it had to. It is possible
that the pakicetid body would be able to handle seawater, but since they
lived in a freshwater ecosystem, they never needed to. Given that at least
some modern artiodactyls can process seawater, the ability to handle life
without a freshwater source may already have existed in the ancestors
of cetaceans.
A few years later, Lois leaves science, and Mark Clementz takes over
the isotope work. Mark is a generation younger than I, very dynamic and
124 | Chapter 9
with a whole new array of sampling and isotope analysis techniques in
his tool box. Most appreciated, from my perspective, is that now we only
need tiny bits of enamel to analyze isotopes. I can barely see it when
Mark drills into a whale tooth. Also, techniques have now improved so
that we can answer much more sophisticated questions. For instance,
Mark is able to use teeth that erupt earlier and later in the life of an ani-
mal and distinguish which of those teeth were formed (in the jaw) before
the animal was born, when it was nursing, and when it became an inde-
pendent feeder—all based on the differences in fractionation of isotopes
at those stages.7
That level of detail will come in handy in studying Ambulocetus,
whose isotope data are intriguing. Its isotope signatures are all over
the place, but are mostly in the freshwater area. That is at odds with
its coastal living environment, where seawater and brackish water
abounded.8 If the ability to handle seawater was not present in the
ancestors of cetaceans, it is possible that Ambulocetus was living at the
shore but had to swim up a river to drink freshwater in order to not
overdose on salt. But other interpretations are equally possible. Maybe
they lived in rivers as juveniles (when their teeth were formed), and
moved out to the coast (where we find their fossils) later on. Or maybe
they were like the alligators on Kiawah Island: they chose the freshwa-
ter habitats in an ecosystem dominated by marine habitats. If those alli-
gators ever fossilize, they will be found among the seashells and shark
teeth, in spite of their habitat. Maybe Ambulocetus did not drink at all,
instead getting all its water from its prey—and those were freshwater
fish or land mammals. Of c
ourse, we are first going to have to find milk
teeth for Ambulocetus, and we haven’t.
walking with ambulocetus
With our isotope work on Ambulocetus in full swing, the whole habitat
issue takes a turn into fiction. The BBC is making a series of documenta-
ries about the evolution of mammals. They call it Walking with Beasts,
and whales play a prominent role. Ambulocetus is seen swimming, walk-
ing, and hunting in one of the episodes. The makers do an excellent job
trying to get the animal right. They send me version after version of little
movie clips of the animal moving across the screen, first as a stick figure,
later more and more realistic, and eventually with fur and a menacing
glare. The makers take my comments seriously: the length of the snout
gets fixed, and so does the flexibility of its spine. The outcome is stunning.
The Ocean Is a Desert | 125
I am fascinated by its looks. It is as if a forty-eight-million-year-old film of
the beast in the wild were discovered. The fascination comes to an abrupt
end when they add the setting of Ambulocetus’ appearance—they put it
in a German fossil site, Messel. In the Eocene, Messel was a near-dead
lake in a forest that belched toxic volcanic fumes. Most animals coming
near it died of the fumes, and fossilization was common because the place
was too toxic for scavengers to live and eat the carcasses. I argue with the
makers of the show. Ambulocetus lived half a world away, on a desert
coast, in waters that brimmed with life, not in a dead and deadly pond in
the German forest. My complaints are acknowledged but rejected. To tell
a good story, the whale needs to paddle happily through that toxic mud
hole in pursuit of rat-sized critters on the forest floor. Don’t believe every-
thing you see on television.
Chapter 10
The Skeleton Puzzle
if looks could kill
Locality 62, Punjab, Pakistan, 1999. Six of us are back at locality 62,
the place where Robert West found the first Pakicetus, digging for more
The Walking Whales Page 19
