but not humans, the end of the soft palate touches a piece of cartilage
that forms a valve (the epiglottis), and that contact helps to prevent chok-
ing. Closing that valve seals off the entry to the trachea (the windpipe)
when the animal swallows, thus avoiding food going into the trachea.
Learning to Swim | 61
In humans, the larynx is located lower in the neck, and the epiglottis
does not reach the soft palate. Lacking the palate–epiglottis seal, humans
choke much more easily than most mammals. However, the increased
space between these structures is important in speech.
The human larynx moved down into the neck in evolution, whereas
in odontocetes it has moved up,23 and now protrudes into the nasopha-
ryngeal duct, making a very tight seal. This causes the air passage and
the food passage to be completely separated. No food or drink can go
up the nose in a dolphin, important for an animal that feeds under-
water. It is possible that the extended hard palate of Ambulocetus also
served to keep food and air passages separate. However, if that was
indeed its function, the mechanism was certainly very different from
the mechanism that modern cetaceans use. More study is needed to
understand the function of the nasopharyngeal duct in Ambulocetus,
and it is not difficult to imagine other functions. Maybe, just like in
humans, the throat was used in making sounds.
Even more puzzling are the bones that surround the throat of Am-
bulocetus. In all mammals, the larynx is supported by some bones and
pieces of cartilage that together are called the hyoid.24 The bones of
the hyoid are not usually preserved in fossils, but in whales they are
unusually large, and hyoid bones are known for a number of extinct
species. In Ambulocetus, the three bones that make up the hyoid were
found still partly in their position with regard to the skull (figure 18).
However, their depth is barely larger than the depth of the nasopha-
ryngeal duct. And yet, all food has to pass between this duct and the
hyoid on its way to the stomach. The space available may just admit
something the size of a golf ball, meaning that Ambulocetus only swal-
lowed things smaller than that. Modern toothed whales do not chew
their food. They tend to swallow large chunks. Killer whales are rou-
tinely found with entire seals in their stomachs, for instance.25 Ambu-
locetus’s mouth and throat hardware clearly worked differently from
modern whales—it ate smaller chunks—but more research is needed
to fully understand the similarities and differences.
Vision and Hearing. The position of the eyes in Ambulocetus is differ-
ent from that in basilosaurids. The eyes of Basilosaurus and Dorudon
are large. They face laterally, and they are located on the side of the
head, under the thick bony shelf called the supraorbital process. In
Ambulocetus, the eyes are also large, but they are perched on top of the
head, close to the midline, and they face partly sideways and partly up.26
That position suggests that Ambulocetus could stay submerged while
lifting just its eyes out of the water to survey its aerial surroundings,
much as alligators do. Ambulocetus clearly had an interest in watching
62 | Chapter 4
Deer
mandibular
Odocoileus
condyle
Pakicetus
mandibular
H-GSP 1694/92034
symphysis
Ambulocetus
mandibular
H-GSP 18507
foramen
Dorudon
Kellogg, 1936
figure 25. The lower jaw of a
deer and some fossil and modern
cetaceans. The size of the
mandibular foramen increases in
whale evolution. This foramen is
Dolphin
involved in sound transmission
Delphinus
mandibular
under water. The mandibular
DO 5837
symphysis
symphysis is very large in late
Eocene whales ( Dorudon). These
drawings are not to the same
scale.
things above the water, and it is possible that the whale found its prey
there.
The shape of the lower jaw of Ambulocetus provides some clues
about the evolution of hearing. In all mammals, there is a small hole in
the back of the lower jaw, the mandibular foramen (figure 25). Through
this foramen travel an artery, nerve, and vein that supply the lower
teeth. Dentists target this nerve with their syringe if they need to numb
the lower teeth of a patient. In most mammals, this foramen is just big
enough to transmit those three structures. Not so in modern dolphins,
other toothed whales, and basilosaurids, where the foramen is very
Learning to Swim | 63
large and houses a fat pad. That fat pad has an important function in
hearing.27 In Ambulocetus, as well as in Himalayacetus, the foramen is
intermediate in size, larger than in land mammals but not the size of
basilosaurids’ and odontocetes’ either.28 That intermediate size suggests
that the sound-transmission mechanism of the ear is evolving in these
whales; this will be further discussed in chapter 11.
Walking and Swimming. Most of the vertebrae of the chest and back
were found for that one individual of Ambulocetus initially found by
Mr. Arif.29 The neck vertebrae for this animal are poorly preserved, and
much of the tail is missing. There are sixteen thoracic vertebrae with
many preserved ribs, eight lumbar vertebrae, and four vertebrae fused
into a sacrum. Many of these were found in articulation, so they were
buried with flesh still holding them together.
These numbers are surprising. In most mammals, as well as birds and
reptiles, the total number of cervical (neck), thoracic (chest), and lum-
bar (back) vertebrae is around twenty-six.30 In birds, there are many
cervical vertebrae and few lumbar vertebrae, but it still adds up to
around twenty-six. In mammals, there are (nearly always) seven cervical
vertebrae, and the thoracic and lumbar numbers vary inversely, making
the total number of presacral vertebrae also add up to around twenty-
six. Ambulocetus’s thirty-one is different, and basilosaurids have even
more. Clearly, cetaceans are altering some of the basic structural designs
of mammals. We will get back to that in chapter 12.
The Ambulocetus skeleton that we found was from a young indi-
vidual. Many of its vertebrae still have areas where growth was taking
place. In life, these areas contained cartilage discs called growth plates
(epiphyseal plates), which disappear after growth ceases in an adult. In
most cetaceans, growth plates disappear first in the neck and tail, and
then move toward the vertebrae m
ore or less in the middle of the ani-
mal. Indeed, this is also the case in Ambulocetus. 31
Finding the sacrum of Ambulocetus was a real treat. It has four fused
vertebrae, with a firm joint for the pelvis, similar to land mammals and
different from other whales, including basilosaurids. The posterior part
of the pelvis, where the hamstring muscles attach, is also large. The same
is true in seals, where those muscles are used in kicking back the legs.
The forelimb was less flexible than in many land mammals: the
radius and ulna (the bones between elbow and wrist) were immovably
locked against each other. That indicates that the animal could not cup
its hand (supinate). There were five fingers, with a normal complement
64 | Chapter 4
The fossil whale
Ambulocetus natans
Foot
Hand
1
2
4
5
5
3
2
4
3
figure 26. The bones of the hand (forefoot) and foot (hind foot) of Ambulocetus
natans (H-GSP 18507). The animal had five fingers (1 is the thumb) but only four
toes (labeled 2 through 5); it did not have a big toe. Gaps indicate bones that were
not discovered.
of metacarpals and phalanges (figure 26). Fingers as well as toes ended
in a low and long hoof, similar to the hoof of a deer. That suggests that
cetaceans were related to hoofed mammals. Flanges on the sides of
Ambulocetus’s phalanges of the foot indicate that the foot may have
had webbed toes.
Habitat and Life History. With only one relatively complete specimen,
not much can be said about the life history of the species. But even
though this was a young individual, there was extreme and unusual
tooth wear, suggesting heavy use. Given that there is a lot of relief
between the front and back of each tooth, and that the tips of the molars
are high, it is unlikely that some very abrasive food caused the wear. It
is more likely that Ambulocetus chewed in a very specialized way, but
we do not know how.
Shells of marine snails and sirenian ribs occur near the site where
Ambulocetus was found, indicating that the ocean was near. Ambuloce-
tus lived a hot climate, in water that varied in salinity.32 However, fresh-
water was not far off, as evidenced by the fossils of land mammals in
Learning to Swim | 65
nearby rocks. It is likely that Ambulocetus lived on the edge between
land and water, as well as on the edge between freshwater and saltwater.
It was a transitional form in more than one way.
ambulocetus and evolution
Ambulocetus is often referred to as a missing link: a critical fossil that
combines features of two groups that it is related to, while not much
else ties those two groups together. From the perspective of creationists,
missing links do not exist. Duane Gish devoted an issue of Creation to
the beast and stated that it is “probably an animal related to seals.”33
With that, he acknowledged its amphibious nature, but ignored the
plethora of cetacean features. Gish claimed that missing links cannot
exist because they would have no hope of survival in between two dif-
ferent environments. When it was first described, Ambulocetus defused
that argument, to the delight of the staunch defenders of evolution.34
Such victories aside, it is important to remain humble about our under-
standing of all evolutionary transitions. As one colleague put it, “Every
missing link that is discovered creates two new ones, one on either side
of discovery.”
Ambulocetus is a critical piece in the puzzle of cetacean origins, but
most of the puzzle cannot be discerned. With this new fossil, many
parts of the body are similar enough to land mammals that detailed
comparisons can be made that may allow us to determine to which
land mammal cetaceans are most closely related. It might contribute
to a controversy that is brewing in the scientific world. Paleontologists
assume that cetaceans are closely related to an extinct group of hoofed
carnivorous mammals called mesonychians,35 but molecular biologists
note many similarities between the DNA and proteins of cetaceans
and those of even-toed ungulates (artiodactyls) in particular hippo-
potamids.36 But Ambulocetus does not resolve the question. More
fossils of the first whales are needed, like those of that enigmatic Paki-
cetus. Pakistan is where both of those were found, making me think
that the place of origin of cetaceans is in that region. I want to go back
there, and find fossils that answer these questions. That is, if I get the
funds, and if it remains safe to travel there. Those are big “ifs.”
Chapter 5
When the Mountains Grew
the high himalayas
Plane Over Pakistan, May 23, 1994. Visits to the Indian subcontinent
are best done between December and April, after it has recovered from
the drenching monsoon rains in fall, but before the summer sun parches
it. This year, I do not follow that recommendation. I arrive on the plains
of Punjab during the mango season—the one benefit of traveling in
May. I have to travel at this time because we want to go to the high
Himalayas, where snow, avalanches, mudslides, and bitter cold make
collecting in winter all but impossible. But the Indus Plain, where we
will soon land, is miserable now. When the British ruled India, they
moved the top level of government into the mountains during summer,
leaving typhus and dysentery to rule the plains. Army and civil-service
officers not at the top level would send their families to small mountain
towns called hill stations, while they stayed behind to manage the native
masses. The hill stations developed an unusual British countryside fla-
vor, with churches and English bungalows on forested slopes. Social life
in these towns was dominated by ladies and children, revolving around
Victorian teas and balls. Young men on leave from their jobs in the heat
or on temporary assignment to the hill station were in high demand at
these parties to combat boredom, feed gossip, and satisfy forbidden pas-
sions. Many hill stations still retain an English flavor in their buildings,
albeit now all inhabitants are dressed in shalwar kameez.
67
68 | Chapter 5
We land in Islamabad before dawn, and as soon as the airplane door
opens, a cocktail of airplane exhaust and tropical heat overwhelms me.
In seconds, sweat soaks my shirt, even though the sun has not cracked
the horizon. I hope to only spend a few days i
n Islamabad, just enough
to get the fossil-collecting gear and group together. My colleague Taseer
Hussain is here already and has organized a red jeep and an old beige
Land Rover for use as vehicles. With Taseer, everything is understated.
He is a gentle man with an ironic smile and a Pakistani lilt.
“Well, Land Rover is old and may not come back, this is why you
have jeep.”
I like the Land Rover. It reminds me of the British nature shows I
used to watch as a kid. I suppose it will go down fighting the harsh
mountain terrain.
“Very good, who all is going to Skardu?”
“Two drivers, Munir and Raza, I believe you know them. Also, head
cook Rookoon, assistant cook, and Mr. Arif.”
“That seems like a lot of people.”
“It is, everyone wants to see Skardu and cool off in mountains.”
I think to myself that I want to be part of a fossil-collecting expedi-
tion, not a sightseeing tour, but I keep my mouth shut.
“You are not coming?”
“No, Karakorum Highway is only for young men like yourself. I will
fly to Skardu after you arrive. You can pick me up at the airport there.”
I consider the “only for young men,” and wonder what he means, but
decide not to ask. Moreover, I have no interest in flying there. The drive
will be spectacular.
“I did not know that you could fly to Skardu.”
“Oh, yes. Flight is very erratic, it only takes off when weather is
good. Landing is very tricky there. Pilot has to circle down to land. Val-
ley is too small for a straight approach.”
OK, so now I am happy not to fly. Skardu is a small town in a valley
on the Indus (figure 1). Much of that river runs in a narrow gorge in the
mountains, but at Skardu the valley broadens and some agriculture is
The Walking Whales Page 10