neck should be. Toward the tail, tiny hind limbs are visible, but they are
too small to bear weight or help in swimming. It has been suggested that
they were used in mating,16 similarly to the claspers that male sharks use
to help hold on to females when they copulate. At the end of the tail, we
find the best evidence confirming our hunch that this was a whale: Basi-
losaurus had a horizontal tailfin, a fluke.
Because the animal is actually more than thirty million years old, we
have only its skeleton and so we do not know whether it had fur, or sparse
hair, or was naked like modern whales. Some scientists have tried to figure
this out by studying modern animals, but the results remain ambiguous.17
Our information of Basilosaurus, Zygorhiza, and their relatives comes
from the fossil skeletons that were mostly found in the deserts of Egypt and
in the southern United States. These rocks were formed between thirty-
four and forty-one million years ago. Both genera are included in the fam-
ily Basilosauridae (basilosaurids in English),18 which is traditionally divided
into two subfamilies: Basilosaurinae, giant, elongated snake-like forms,
and Dorudontinae (figure 8), shorter forms that somewhat resemble a dol-
phin in body shape.19 In everything except their vertebrae and body shape,
the two groups are very similar. Complete skeletons of the basilosaurine
Basilosaurus show that the animal was about eighteen meters (sixty feet)
long, whereas dorudontines, for instance Dorudon, were around a quarter
of that (figure 9).20 Basilosaurids have been discovered in many places all
over the world and were probably distributed worldwide (figure 10).
figure 9. Skeletons of two fossil basilosaurid whales: the large Basilosaurus and the
much smaller Dorudon. The picture of Dorudon is repeated in the upper right corner, but now at the same scale as Basilosaurus, to show the great difference in size . After Kellogg (1936), Gingerich et al. (1990), and Uhen (2004).
figure 10. Map of the world forty-five million years ago (the Eocene period), with
places where fossils of basilosaurid whales and protocetid whales (discussed in
chapter 12) have been found. Base map from http://www.searchanddiscovery.com
/documents/2010/30109andrus/images/fig02lg.jpg and data points from http://
fossilworks.org where the cetacean sections are compiled and edited by Mark Uhen; he
reports that the record from Antarctica is ambiguous.
Fish, Mammal, or Dinosaur? | 21
Feeding and Diet. If our captive basilosaurid opened its mouth, it
would be immediately clear that it was not like any other living whale.
Most modern toothed whales have teeth that are simple pronged stab-
bers—think of the peg-like teeth of killer whales—with little variation
across the tooth row or between upper and lower teeth. This similarity
in tooth shape is called homodonty ( homoios is Greek for “similar to”).
But basilosaurids had more complex teeth that differ from front to back
in the mouth, like humans and most other mammals. This is called het-
erodonty ( hetero, Greek for “the other, different”). In the front, long
and sturdy pointed teeth would be visible, whereas in the back, each
tooth would have multiple bumps (or cusps, as paleontologists call
them; figure 11).
Teeth and Paleontology
Teeth are very important to mammal paleontologists, because they are
the elements that are most commonly preserved and because they are
highly characteristic in different species. One tooth is often enough to
identify a species. Owen identified Basilosaurus as a mammal based on
a few teeth. Most mammals have four different kinds of teeth in each
jaw quadrant—left and right upper, and left and right lower (figure
11). Think of your own teeth. From front to back, humans and most
other mammals have incisors, canine, premolars (bicuspids), and mo-
lars. Primitive placental mammals, such as moles, have three incisors,
one canine, four premolars, and three molars on both left and right in
both the upper and lower jaw. Paleontologists express that as a dental
formula 3.1.4.3/3.1.4.3, where half of the upper jaw is shown in front
of the slash, and half of the lower jaw behind the slash. The dental
formula is very stable within a species, but can vary greatly from the
primitive placental count. In mice, for instance, the dental formula is
1.0.0.3/1.0.0.3. In humans, it is 2.1.2.3/2.1.2.3.21 Often the upper and
lower dental formulas are not the same. Basilosaurus is an example:
3.1.4.2/3.1.4.3. Thus, it has two upper molars but three lower molars.
Throughout evolution, many mammalian groups independently have
reduced the number of teeth from that original number, an important
trend that we will return to in chapter 15.
Most mammals have relatively simple incisors (see Box), including
Basilosaurus, which has a simple pointed prong with one root.22
In most mammals the canine is bigger than the incisors, but in Basilo-
saurus it is similar to the incisors . The premolars of Basilosaurus
figure 11. The adult dentition of the basilosaurid whale Dorudon and the mole
Talpa, at very different scales . For each species there are side views of the left teeth (lateral views, middle drawings) and views of the chewing surface (occlusal views,
top and bottom). Talpa has a dentition that is characteristic of basal placental
mammals, including early ancestors of whales: three incisors, one canine, four
premolars, and three molars on left and right side, with each of the molars showing
complex morphology of lows and highs (cusps). Dorudon teeth are simpler in
shape: the valleys between cusps have disappeared, and there are only two upper
molars.
Fish, Mammal, or Dinosaur? | 23
increase in numbers of cusps from front to back. Molars are also
complex: each molar crown has a row of pointed cusps, and each molar
has two roots.
A number of basilosaurid fossils are from young individuals who still
had their milk, or baby, teeth. That is surprising, because modern ceta-
ceans do not have a milk dentition: the first generation of teeth that
erupt in a baby dolphin are the only teeth it will ever have. Thus, that
too has changed in cetacean evolution.
The entire dental ensemble in basilosaurids was powered by very
significant jaw muscles that covered the entire top of the skull, arising
from a large crest on top of the head, the sagittal crest. No doubt about
it: basilosaurids could bite hard. What did they eat? Some of the articu-
lated skeletons show an accumulation of fish bones located in the area
where the stomach would have been, and these have been interpreted as
stomach contents.23 Microscopic scratches on the teeth also look like
the scratches in modern fish-eaters like seals,24 so it seems that basilo-
saurids ate fish. One specimen has shark teeth in its belly, showing that
small sharks at least were not a match for the King Lizard of Cape Cod.
Also, there are tooth marks on the skull of a juvenile Dorudon that
match the distance between teeth of Basilosaurus, suggesting that one
basilosaurid ate others.25
Brain. Since the 1800s, a valley
in Egypt called Zeuglodon Valley, or
Wadi al-Hitan (Valley of the Whales), has yielded a wealth of basilosau-
rid skeletons. There, fierce winds scour the surface and carry away the
sediment, exposing fossils. The exposure is temporary: eventually the
fossil bone too is devoured by the wind, turned to powder, and blown
away. Cavities in the fossils, such as the cavity in the skull where the
brain used to be, are filled with fine sediment which is harder than the
bone. So as the bone erodes away, the filled cavities remain. As a result,
many fossils from this area are endocasts: lumps of hard sediment that
preserve the shape of the cavity they once filled. Not only bone leaves
impressions on the sediment. Many of the soft structures inside the
braincase leave impressions on the bone too, making it possible to learn
about anatomy that itself does not fossilize. Researchers have described
cranial endocasts of basilosaurids in detail, and some were even named
as separate species by overzealous paleontologists.26 Endocasts can be
used to estimate brain size, too. From this, it is clear that basilosaurids
had tiny brains, much smaller than even those of modern cetaceans with
small brains, such as bowhead whales (see Box).27
Brain Size
The volume of endocast of the cranial cavity (where the brain sits)
can be measured and used as an estimate of brain size, and provide
some indications of an ancient animal’s intelligence. The cranial cavity
contains several organs besides the brain, such as arteries, nerves, and
the membranes that protect the brain (the meninges). Those structures
often also leave impressions in an endocast, and those impressions are
often not clearly distinct from the brain impression itself. So, measures
of endocranial volume are an overestimate of actual brain volume in
vertebrates. In a horse, for instance, 94 percent of the cranial cavity is
filled by the brain.28 Matters are worse in cetaceans, because, at least
in the modern species, a large mass of veins envelops the brain. Such
masses are called retia mirabilia (plural of rete mirabile or “wonder
net”). Endocranial size has been estimated by dunking endocasts in
water and seeing how much water is displaced, or, in modern times,
by using CT-scan technology.29 This has given us a good idea of how
endocranial size changed in cetacean evolution. However, brain size
may not follow this pattern, because the size of the rete may also have
changed in evolution. Actual measurements on the skull and brain in
a bowhead whale, a modern baleen whale, indicate that only 35 to 41
percent of the cranial cavity is filled with brain in this species.30 That
makes it difficult to tease apart the pattern of brain evolution from
that of endocast evolution, although some broad patterns emerge.
Brain size is most meaningful when it is scaled with body size.
Larger animals have larger brains simply because a larger body needs
a larger brain to operate it. So if we are interested in studying brain
size, we need to correct for body size. To make that comparison, sci-
entists calculate a ratio called the encephalization quotient (EQ).31 At
any one body size, a mammal with an average-sized brain has an EQ
of 1, an animal with a larger-than-average brain has an EQ greater
than 1, and a smaller-than-average brain an EQ smaller than 1. Cats,
for instance, have an EQ of 1; they have an average brain size for their
body weight. Horses have an EQ of 0.9, and it is 2.5 in chimpanzees.
Humans have the highest EQ on the planet: over 7. In the bowhead
whale, the EQ is 0.4,32 similar to that of a rabbit. The point has been
made that this number is misleading since fat makes up 40 to 50 per-
cent of the weight of a whale and fat needs less brain tissue to operate
it than other tissues do, thus artificially lowering the EQ. If we correct
the body-weight value by ignoring the weight caused by fat altogether,
the recalculated EQ for a bowhead is 0.6, still low.
Fish, Mammal, or Dinosaur? | 25
Vision, Smell, and Hearing. If you were watching our captive basilo-
saurid come up to breathe, you would probably notice that its nose
opening was halfway between the tip of the snout and the eyes. It is
unclear why the opening is so far back, although scientists have specu-
lated that underwater life favored this position. After all, most living
whales have blowholes far back on their heads, and can breathe while
just exposing the smallest part of their body. But most vertebrates that
live underwater have their nose opening at the tip of the snout, for
instance seals, manatees, hippos, muskrats, and even underwater preda-
tors such as crocodiles, sea snakes, and sperm whales. There may be
more to the evolution of the blowhole than just underwater living. The
shifted position of the nasal opening certainly caused there to be less
room for tissues involved with the sense of smell, but from the bones of
the nose, it is clear that basilosaurids had a sense of smell.
The eyes of basilosaurids were directed toward the sides; they are
located under a broad shelf in the skull, called the supraorbital shelf of
the frontal bone. Their visual field is thus mostly directed toward the
side, and this suggests that they were hunting prey underwater, which is
consistent with what we know about their diet.
We know a lot about basilosaurid hearing, because many of their
fossils are very well preserved and include such rarely preserved pieces
as the ear ossicles (figure 3). Their ear ossicles are very similar to those
of modern whales,33 suggesting that, like modern whales, basilosaurids
had keen hearing underwater (see chapter 11).
Walking and Swimming. With their serpentine body and tiny hind limbs,
basilosaurids could not get around on land. Their home was the ocean—
they are obligate aquatic animals. The vertebral column reveals that basi-
losaurids are mammals, not dinosaurs or fish. There are seven neck verte-
brae, a number typical of mammals from giraffe to human. In basilosaurids,
as well as modern whales, these vertebrae are very short; as a result, the
shoulders are so close to the head that the neck disappears. Then there are
seventeen thoracic (back) vertebrae, each of which carries a pair of ribs.
Those ribs are interesting.34 The part that reaches to the chest side of the
animal (the ventral part) is very heavy and dense, a condition called oste-
osclerosis ( os means bone in Latin; scleros means hard). This part of the
ribs is also a bit thicker than the rest, which is called pachyostosis ( pachus
means fat in Greek). Such extra weight in the skeleton is important
in some marine mammals because it provides ballast that allows them
to stay submerged.35 But these features are not usually present in fast
26 | Chapter 2
predators such as many modern whales and dolphins, and compared to
other mammals, basilosaurid bones are just mildly osteosclerotic.36 It
remains likely that dorudontines, like the dolphins that are similar in
; shape, were pursuit predators of fast-moving fishy prey.
However, the difference from modern whales begs for an explana-
tion. Why do basilosaurids’ ribs weigh them down? The position of the
pachyostosis, on the ventral side of the rib, is suggestive. Perhaps, by
concentrating weight on the belly side, the weight helped to keep the
animal from going belly-up during swimming. The dorsal fin of modern
whales—which is not made of bone and so wouldn’t fossilize—helps
with that job, acting like the keel on a ship in preventing rolling. We
don’t know if basilosaurids had a dorsal fin, but it may be that they
lacked one and that the pachyostosis was an anti-rolling device.
Behind the thoracic vertebrae, Dorudon has forty-one vertebrae that
change only very gradually in shape and size, reaching the tip of the tail.
In land mammals, these vertebrae are divided into lumbar, sacral, and
caudal vertebrae, and vary greatly in shape. The sacral vertebrae of land
mammals fuse together to form a composite bone, the sacrum, that
transmits weight to the pelvis and from there to the hind limbs.37
In modern whales, no vertebrae in this region fuse. Paleontologist Mark
Uhen of George Mason University studied Dorudon in detail and found
that even though there is no sacrum, vertebrae 17 through 20, behind the
thoracic vertebrae, are different. These vertebrae have projections (trans-
verse processes) that are much thicker than those of adjacent vertebrae. In
land mammals, these transverse processes on the sacrum articulate with
the pelvis, connecting hind limb to vertebral column.38 It is likely that these
vertebrae represent the sacrum. That allows us to identify the sacral verte-
brae in this fossil whale, and the lumbar vertebrae in front of it. And that
shows that there are many more lumbar vertebrae in basilosaurids than in
The Walking Whales Page 4
