most land mammals, although it is not clear what the function of this is.
In Basilosaurus, the vertebral numbers are similar to those of Doru-
don but their shape is different. In Basilosaurus, the centrum of the lumbar and caudal vertebrae is enormous, like a massive cylinder larger
than a can of paint; the vertebral arch is very small in comparison (fig-
ure 6). This allows for great mobility in all directions, as expected if the
animal was snake-like in its locomotion.39
The triangular fin at the end of the tail of a modern whale is called
the fluke, and basilosaurids had one too. In modern whales, the fluke is
the shape of a symmetrical triangle.40 Internally, the fluke consists of a
row of tail vertebrae that runs down the center, with thick triangular
Fish, Mammal, or Dinosaur? | 27
figure 12. Vertebral shape of two seacows (sirenians: manatee and dugong) and a
cetacean (the modern dolphin Delphinus). Note how in the animals with flukes (dugong
and dolphin) the vertebrae are abruptly narrower at the point where the fluke is
attached to the body (the peduncle). Flukes do not fossilize, but vertebral shape change
can be used to infer the presence of a fluke in fossils. After Buchholtz (1998).
pads of connective tissue and skin making the extensions toward the
sides. There is no bone in the triangular side flaps, so it does not fossilize
and we don’t have a preserved basilosaurid fluke. But we know they had
one, because tail vertebrae in an animal with flukes are different from
those in a non-fluked one. The massive part of these vertebrae (the cen-
trum) has different proportions from the vertebrae in front and behind
it (figure 12). The base of the fluke is called the peduncle, and here the
centrum of the vertebrae is higher than it is wide, whereas further to the
front and further to the back, these proportions are reversed.
In addition to that difference, the vertebra located right at the base of
the fluke has convex anterior and posterior surfaces, and is called the
ball vertebra. Basilosaurines and dorudontines had both these features,
and so they had a fluke. It is likely that the dorudontines used their fluke
to propel themselves during swimming, similarly to modern cetaceans.
On the other hand, this is less clear in Basilosaurus. It is often said that
these whales did not use their fluke for propulsion but swam by means
of serpentine movements of their spine, as helped by the very flexible
28 | Chapter 2
figure 13. Hand of human and flipper of two whales. Humans show the
ancestral pattern: five fingers, each with a metacarpal and three phalanges (two in
the thumb). The fossil whale Dorudon has a single phalanx in each finger, or
possibly one more phalanx in some individuals. Modern whales, such as the right
whale, often have more phalanges, and the fingers are always embedded in a
flipper.
spine. Body shape and swimming methods have been studied a lot in
fish,41 and Basilosaurus’s body has been compared to a giant eel, although
some basilosaurid specialists, such as Uhen, doubt this interpretation.
Living cetaceans use their forelimbs mostly for steering, balance, and
in starting and stopping; these limbs barely help in propulsion. Basilo-
saurids may have done the same. For the wrist and hand, only a few
fragments were known. Mark Uhen described the forelimb for Doru-
don and he found that the shoulder joint was relatively mobile, similar
to that of a modern cetacean.42 The elbow is not mobile in modern ceta-
ceans, whereas that of basilosaurids allowed some bending and stretch-
ing. Movements at the wrist were just about impossible in basilosaurids,
as in modern cetaceans. The fingers allowed some movements, unlike
most modern cetaceans.43 By comparing to other marine mammals,
Uhen concluded that basilosaurid hands (or forefeet) were embedded in
a stiff paddle, a flipper, just like modern cetaceans (figure 13). Inside
that flipper were five bony fingers, as in most modern cetaceans as well
as most other mammals. In humans, the palm of the hand holds five
bones called metacarpals; each of those is followed by three bones
(phalanges) that make up the segments of the fingers (the thumb has just
two). In basilosaurids, the fingers appear to have had a metacarpal and
just one phalanx, although it is not clear whether these bones were miss-
ing in the living animal or were lost during fossilization. If they were
indeed absent in the living animal, it would mean that basilosaurids
Fish, Mammal, or Dinosaur? | 29
figure 14. Sacrum, pelvis (innominate) and femur (thigh bone) of two land
mammals, two fossil whales, and one modern whale. In most mammals, the sacrum
consists of multiple vertebrae, one of which articulates with the pelvis, and the
pelvis has a mobile joint with the femur (human, dog, and Ambulocetus, discussed
in chapter 4). In Basilosaurus and all modern cetaceans, the connection to the
vertebral column is lost. However, Basilosaurus still retains a joint between pelvis
and femur.
lost, in evolution, two phalanges per finger compared to their ancestors,
which would have interesting implications for their embryonic develop-
ment (see chapter 13). It would also be surprising from an evolutionary
perspective, since most modern whales have three or more phalanges.
There are too many “ifs” here, but if some basilosaurids are ancestral to
modern whales, the number of phalanges for the fingers would have
gone from three (in the land ancestors of whales) to one or two (in basi-
losaurids) and back to three or more again (in most modern whales).
More fossils are needed to clear up this issue.
Basilosaurus has tiny hind limbs, a few feet long, attached to a sixty-
foot-long body. Although no complete hind limbs are known for basilo-
saurids, there are enough fossils to indicate that the other basilosaurids
had hind limbs similar to Basilosaurus. The hind limb is attached to a
pelvis, which, in land mammals, articulates with the sacrum (figure 14).
To understand the hind limb of basilosaurids, it is useful to first consider
the hind limb in modern cetaceans. Although the number and size of the
hind-limb bones in modern cetaceans varies among species, it does not
30 | Chapter 2
protrude from the body in any modern species: all bones are embedded
in the wall of the a
bdomen (although we’ll get to some exceptions in
chapter 12). Bowhead whales have more parts to their hind limbs than
most other modern cetaceans, although they vary in size among indi-
viduals. In bowheads, there is always a pelvis and femur, a cartilaginous
or bony tibia, and sometimes even a bony metatarsal. Sometimes, there
is a real synovial joint (a joint with lubricating fluid, like all of the highly
mobile joints in the body).44 The left and right pelvis of modern ceta-
ceans do not articulate with each other, and they also do not articulate
with the sacrum (figure 15). In many other modern cetaceans, there are
no hind-limb bones, and the pelvis is a simple prong-shaped bone.45
Although it is not involved in locomotion, the pelvis of modern ceta-
ceans does have a function. In the male, the pelvis anchors the muscles
to the penis and to the abdominal muscles,46 and muscles extend from
these bones to the genitals in the female too.
The pelvis and femur of Basilosaurus were first described in 1900.47
That fossil (figure 14) shows that there was a synovial joint between
pelvis and femur and foramen behind it, like in land mammals, and
unlike (nearly all) modern whales. Those features make it possible to
determine how it was oriented in the body. One end of the bone is
bumpy in texture, and has been interpreted as the point where left and
right pelvis attach to each other in the body’s midline (the pubic sym-
physis). However, modern bowhead whales have a similar textured area
where the penis is anchored. It is likely that this is similar in basilosau-
rids, and that left and right pelvis of basilosaurids did not articulate
with each other or with the sacrum. In shape of the pelvis, basilosaurids
may be closer to modern whales than to other Eocene whales.
Most of the remainder of the foot is known from Basilosaurus from
Egypt. Basilosaurus had a mobile knee with patella (kneecap), but the
ankle consisted mostly of immobile and fused bones. The foot had three
toes, and instead of a metatarsal and three phalanges, Basilosaurus had
a metatarsal and just two phalanges, and those two fused into a single
unit. Clearly, this animal could not bend its toes.
Habitat and Life History. We know little about basilosaurid social
behavior. Some clues can come from the relative sizes of males and
females. In some mammals (including many seals and sea lions, for
instance, as well as gorillas), males are larger than females. Such a dif-
ference in size between the sexes usually occurs when males mate with
multiple females each year (harems). In other marine mammals, most
pelvis
aceta-
femur
pelvis
bulum
femur
pelvis
femur
tibia
tibia
metatarsal?
5 cm
92B20
92B15
92B21
right pelvis
right femur
tibia
(Cartilage)
corpus
penis
spon-
giosum
corpus
pelvis
caver-
ANTERIOR
nosum
06B4
98B5
figure 15. Pelvis and hind-limb bones in four modern bowhead whale individuals,
showing the great variation in size and shape (the scale is the same for all four), with
some having an acetabulum (joint for femur) and others a tibia and even a possible
metatarsal. The diagram shows how these bones are oriented in a male bowhead whale
as seen from the top (dorsal view).
32 | Chapter 2
figure 16. The geological time scale with some important events. Change in
temperature patterns from Zachos et al. (2001); diversity of whales from F. G. Marx and
M. D. Uhen, “Climate, Critters, and Cetaceans: Cenozoic Drivers of the Evolution of
Modern Whales,” (2010) Science 327 (2010): 993–96.
baleen whales for instance, females are larger than males, and in these
species, males do not maintain a harem. The fossil record gives no indi-
cation that basilosaurid males looked different from females, so there
were probably no basilosaurid harems.
Most specimens of basilosaurids have been found in rocks that indicate
that they lived in a shallow sea,48 but some species apparently preferred
specific environments. For instance, the dorudontine Saghacetus is mostly
found in sediments that indicate lagoons, whereas Basilosaurus is found in
rocks that were formed in open water, away from the beach. Basilosaurids
have been found in most oceans (figure 10), suggesting that they swam
well enough to cross large seas. The climate in the time that basilosaurids
lived, the late Eocene, was warm. The poles were bare of ice caps, and the
temperature gradient from poles to equator was not nearly as pronounced
as it is now. Near the end of the basilosaurids’ reign, the planet changed
(figure 16).49 The continents shifted, which in turn transformed the
oceanic currents, interrupting the mixing of equatorial and polar waters.
As a result, the poles cooled, and at the end of the Eocene, Antarctica
began to freeze over. Basilosaurids may have liked the more evenly warmed
Fish, Mammal, or Dinosaur? | 33
waters across the globe that occurred in the Eocene, and may have been
unprepared for the sudden climatic cooling. Or they may have been out-
competed by the new whales that were starting to show up on our planet,
the ancestors of the modern toothed whales and baleen whales.
basilosaurids and evolution
Basilosaurids are impressive cetaceans, resembling modern whales in
many respects with their involucrum, flippers, and fluke. In some respects,
they are intermediate between land mammals and modern whales. For
example, the nasal opening is close to the forehead, unlike in land mam-
mals, and the hind limbs were still present even though they were of no use
in locomotion. The dentition is reminiscent of their land-mammal ances-
tors. For sci
entists, basilosaurids are intermediates: they are evidence that
whales descended from terrestrial mammals. But basilosaurids look too
much like modern whales to help us understand how that dramatic transi-
tion from land to sea happened. And they don’t retain enough ancestral
features to reveal just who their completely terrestrial ancestors were.
The paucity of the fossil record was fodder for those who doubt that
evolution occurred and adhere to a Biblical account of the planet’s his-
tory. Given the gap between four-footed mammals and basilosaurids,
creationists pounced on whales as an example of the impossibility of
evolution. Following up on the trouble Darwin had with whale origins,
creationists claimed that no intermediates would ever be discovered.
Alan Haywood wrote in 1985:
Darwinists rarely mention the whale because it presents them with one of
their most insoluble problems. They believe that somehow a whale must
have evolved from an ordinary land-dwelling animal which took to the seas
and lost its legs. . . . A land mammal that was in the process of becoming a
whale would fall between two stools—it would not be fitted for life on land
or at sea, and would have no hope of survival. 50
For more than 150 years, basilosaurids were our best clue to what
ancient whales looked like. In the early 1980s, West and Gingerich pro-
posed that the Pakistani whales they found were older than basilosau-
rids and much closer to the land ancestors. But these Pakistani fossils
were frustratingly incomplete. The new Pakistani incus suggested an
intermediate condition for yet another organ system and added to the
intrigue. But did it confirm the geographic region that should be scruti-
nized? Were there really undiscovered fossils, buried in Pakistan, that
could bridge this gap?
Chapter 3
A Whale with Legs
the black and white hills
Punjab, Pakistan, December 1991. Last year’s ill-fated field trip to
Pakistan has left me poor, so I can only afford to go alone to Pakistan
this time. There, Mr. Arif and I set out to do fieldwork in a blue Isuzu
pickup truck. That car was new in 1984, on my first trip to Pakistan.
The Walking Whales Page 5